diff --git a/bc/prepare_groceries/pick/all/config.yml b/bc/prepare_groceries/pick/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..b8c298a006c71059beadf4bd8139edda22fe2f60 --- /dev/null +++ b/bc/prepare_groceries/pick/all/config.yml @@ -0,0 +1,46 @@ +seed: 0 +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/all.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/prepare_groceries/pick + max_cache_size: 300000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/prepare_groceries/pick/all/policy.pt" diff --git a/bc/prepare_groceries/pick/all/policy.pt b/bc/prepare_groceries/pick/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..2127abc2794ada668669015b31afe07940223af3 --- /dev/null +++ b/bc/prepare_groceries/pick/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b90849268bcaefe1fa1c1abd79f9e259a48f42617bc152bdeb8943dbd39a45c +size 238201234 diff --git a/bc/prepare_groceries/place/all/config.yml b/bc/prepare_groceries/place/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..739875b497ec8e046e9eab9660498056bfeece96 --- /dev/null +++ b/bc/prepare_groceries/place/all/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/all.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/prepare_groceries/place + max_cache_size: 300000 + trajs_per_obj: all + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/prepare_groceries/place/all/policy.pt" diff --git a/bc/prepare_groceries/place/all/policy.pt b/bc/prepare_groceries/place/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..1767e5761849d29e20498a2ce67f68a36f1f1fa5 --- /dev/null +++ b/bc/prepare_groceries/place/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4bc570e9aad369d26c62d69c5e441f7a46fc45ec20221c5c6f3d11708a49f6e +size 238201234 diff --git a/bc/set_table/close/fridge/config.yml b/bc/set_table/close/fridge/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..11a4ab1406dd09fb52d5161404eacd18b320c4d2 --- /dev/null +++ b/bc/set_table/close/fridge/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: CloseSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/close/train/fridge.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/close/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/set_table/close/fridge.h5 + max_cache_size: 300000 + trajs_per_obj: 1000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-close-fridge + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/set_table/close/fridge/policy.pt" diff --git a/bc/set_table/close/fridge/policy.pt b/bc/set_table/close/fridge/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..e2fad3d8ddfca989a0548767c30d814afebf74b2 --- /dev/null +++ b/bc/set_table/close/fridge/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4e6802c0b767f70b819f74eaf1b063d028e0592cf6ad0cf47739439bf2f1a3f +size 238200850 diff --git a/bc/set_table/close/kitchen_counter/config.yml b/bc/set_table/close/kitchen_counter/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..0f7e5a5fb62ed1ccd74552501e12af11b69a6a11 --- /dev/null +++ b/bc/set_table/close/kitchen_counter/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: CloseSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/close/train/kitchen_counter.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/close/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/set_table/close/kitchen_counter.h5 + max_cache_size: 300000 + trajs_per_obj: 1000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-close-kitchen_counter + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/set_table/close/kitchen_counter/policy.pt" diff --git a/bc/set_table/close/kitchen_counter/policy.pt b/bc/set_table/close/kitchen_counter/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..15ace024a11daa7030c2129c8c4024ae4982f1d5 --- /dev/null +++ b/bc/set_table/close/kitchen_counter/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9cc31e96c04a6e5ffc1f698c9c55e17f2ca136db1865085df71520ec4a59025a +size 238200850 diff --git a/bc/set_table/open/fridge/config.yml b/bc/set_table/open/fridge/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..822ae9f2fc0e54f5f3daaabca994da08023ede4e --- /dev/null +++ b/bc/set_table/open/fridge/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: OpenSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/open/train/fridge.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/open/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/set_table/open/fridge.h5 + max_cache_size: 300000 + trajs_per_obj: 1000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-close-fridge + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/set_table/close/fridge/policy.pt" diff --git a/bc/set_table/open/fridge/policy.pt b/bc/set_table/open/fridge/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..f2fbc583a3cb9de072f6296e786abbf68c80c93f --- /dev/null +++ b/bc/set_table/open/fridge/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:80f614cdfbef42079f9f4254224484d71d49057801182417628d181646cefe38 +size 238200850 diff --git a/bc/set_table/open/kitchen_counter/config.yml b/bc/set_table/open/kitchen_counter/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..a25f551528738065b5e6d265b46e79b9959a7abc --- /dev/null +++ b/bc/set_table/open/kitchen_counter/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: OpenSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/open/train/kitchen_counter.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/open/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/set_table/open/kitchen_counter.h5 + max_cache_size: 300000 + trajs_per_obj: 1000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-close-kitchen_counter + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/set_table/close/kitchen_counter/policy.pt" diff --git a/bc/set_table/open/kitchen_counter/policy.pt b/bc/set_table/open/kitchen_counter/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..fe534bdf1fbbe3529db075f468c9687e9a00b60f --- /dev/null +++ b/bc/set_table/open/kitchen_counter/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4dc77a7e29552c1d407a38bd608173d12d40da3dd35eec934b537a9024b8c21e +size 238200850 diff --git a/bc/set_table/pick/all/config.yml b/bc/set_table/pick/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..e0e3200336ceaf57a90c382e25ccd7bced0dcc21 --- /dev/null +++ b/bc/set_table/pick/all/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/all.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/set_table/pick + max_cache_size: 300000 + trajs_per_obj: all + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-pick-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/set_table/pick/all/policy.pt" diff --git a/bc/set_table/pick/all/policy.pt b/bc/set_table/pick/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..c5bb1e29ea388eecfe62a292cb02c2091572f078 --- /dev/null +++ b/bc/set_table/pick/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a3593b72dfc3c265fb8a294b6f3bb79c86472d4efada6437d50c66daee7cbd58 +size 238200850 diff --git a/bc/set_table/place/all/config.yml b/bc/set_table/place/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..62e74a68e717ec8acb15931050bb0f43a4d77f26 --- /dev/null +++ b/bc/set_table/place/all/config.yml @@ -0,0 +1,47 @@ +seed: 0 +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/all.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/set_table/place + max_cache_size: 300000 + trajs_per_obj: all + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-place-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/set_table/place/all/policy.pt" diff --git a/bc/set_table/place/all/policy.pt b/bc/set_table/place/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..363d7930789c600415a7302d66782b29462363b1 --- /dev/null +++ b/bc/set_table/place/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:39cb5b0a2000c8184a5e4a25cda28e60602734a81e05be972a18f27f911bda2a +size 238200850 diff --git a/bc/tidy_house/pick/all/config.yml b/bc/tidy_house/pick/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..2e3a3e6926b3fdf5e75216f1cc360dad7ccfe25d --- /dev/null +++ b/bc/tidy_house/pick/all/config.yml @@ -0,0 +1,46 @@ +seed: 0 +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/all.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 25 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/tidy_house/pick + max_cache_size: 300000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/tidy_house/pick/all/policy.pt" diff --git a/bc/tidy_house/pick/all/policy.pt b/bc/tidy_house/pick/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..bae87c7917733f8c17783aa0c86c8f068444abaf --- /dev/null +++ b/bc/tidy_house/pick/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:250f92d9a0d6c42f78ba85ac0cab740eee1f80619f6f6fbf39b7ad8aa676d3c8 +size 238201234 diff --git a/bc/tidy_house/place/all/config.yml b/bc/tidy_house/place/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..8e24b9136ef5394997595908bd362e653a495bdb --- /dev/null +++ b/bc/tidy_house/place/all/config.yml @@ -0,0 +1,46 @@ +seed: 0 +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 252 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 1 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/all.json + spawn_data_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 1 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: bc + lr: 0.0003 + batch_size: 512 + epochs: 10 + eval_freq: 1 + log_freq: 1 + save_freq: 1 + save_backup_ckpts: false + data_dir_fp: ~/.maniskill/data/scene_datasets/replica_cad_dataset/rearrange-dataset/tidy_house/place + max_cache_size: 300000 + torch_deterministic: true + eval_episodes: 252 + num_eval_envs: 252 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/bc/tidy_house/place/all/policy.pt" diff --git a/bc/tidy_house/place/all/policy.pt b/bc/tidy_house/place/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..b9e75f54c828546090db737c81443f3bd917dcec --- /dev/null +++ b/bc/tidy_house/place/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e947b7ab786abd6817536bd7af53897f97c89d1214d06ac0c025df5c915b0f2 +size 238200850 diff --git a/rl/prepare_groceries/pick/002_master_chef_can/config.yml b/rl/prepare_groceries/pick/002_master_chef_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..c4a235b4bf482459d177f31295600516ce8d4403 --- /dev/null +++ b/rl/prepare_groceries/pick/002_master_chef_can/config.yml @@ -0,0 +1,110 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-002_master_chef_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/002_master_chef_can/policy.pt" diff --git a/rl/prepare_groceries/pick/002_master_chef_can/policy.pt b/rl/prepare_groceries/pick/002_master_chef_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..85c4f63edfa56096e6c6944d12d8958394cfcdc0 --- /dev/null +++ b/rl/prepare_groceries/pick/002_master_chef_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb338580bfb74b7d50796705f8f5f5caccfe6d5e3a22a57485b5ccab9690ef98 +size 55487432 diff --git a/rl/prepare_groceries/pick/003_cracker_box/config.yml b/rl/prepare_groceries/pick/003_cracker_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..988437d658fe5177af5a03615e1f4051803232b4 --- /dev/null +++ b/rl/prepare_groceries/pick/003_cracker_box/config.yml @@ -0,0 +1,110 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-003_cracker_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/003_cracker_box/policy.pt" diff --git a/rl/prepare_groceries/pick/003_cracker_box/policy.pt b/rl/prepare_groceries/pick/003_cracker_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..1193d5ec375babea5161f51dae4a442836fa956b --- /dev/null +++ b/rl/prepare_groceries/pick/003_cracker_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d11038630d4ed9ccc9de2c8bc1522b8e104b8180c2c87cdf82e7cc10094c88c +size 55487432 diff --git a/rl/prepare_groceries/pick/004_sugar_box/config.yml b/rl/prepare_groceries/pick/004_sugar_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..f93afda9db82fa40f48d79b80b0bd544f8678f4e --- /dev/null +++ b/rl/prepare_groceries/pick/004_sugar_box/config.yml @@ -0,0 +1,111 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-004_sugar_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/004_sugar_box/policy.pt" diff --git a/rl/prepare_groceries/pick/004_sugar_box/policy.pt b/rl/prepare_groceries/pick/004_sugar_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..1b290d78e69e018a36622999da151d16b51299b2 --- /dev/null +++ b/rl/prepare_groceries/pick/004_sugar_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc935696ffe962efbe8e418312082245a90f43abd4a9aecaed1a130132478576 +size 55487432 diff --git a/rl/prepare_groceries/pick/005_tomato_soup_can/config.yml b/rl/prepare_groceries/pick/005_tomato_soup_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..da933f98cb44d2dd861848ae20bcdd471484cc31 --- /dev/null +++ b/rl/prepare_groceries/pick/005_tomato_soup_can/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-005_tomato_soup_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/005_tomato_soup_can/policy.pt" diff --git a/rl/prepare_groceries/pick/005_tomato_soup_can/policy.pt b/rl/prepare_groceries/pick/005_tomato_soup_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..0939fe46ad225acd2c16089f4289d425b6f6d349 --- /dev/null +++ b/rl/prepare_groceries/pick/005_tomato_soup_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77288352508ea9e9aab0499c0a8a1bfd680cba704c98d36d8bbfbc84be8f7c4c +size 55487706 diff --git a/rl/prepare_groceries/pick/007_tuna_fish_can/config.yml b/rl/prepare_groceries/pick/007_tuna_fish_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..e2e9556002e83656a9f46801fc1d1e6fd715decb --- /dev/null +++ b/rl/prepare_groceries/pick/007_tuna_fish_can/config.yml @@ -0,0 +1,111 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-007_tuna_fish_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/007_tuna_fish_can/policy.pt" diff --git a/rl/prepare_groceries/pick/007_tuna_fish_can/policy.pt b/rl/prepare_groceries/pick/007_tuna_fish_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..a7b0aacdbea22d17b5294356dccdf01c9bb3d6d7 --- /dev/null +++ b/rl/prepare_groceries/pick/007_tuna_fish_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caf45664947dc617cc252782f7883395cc1b50472c313864220444ffe9eb2701 +size 55487432 diff --git a/rl/prepare_groceries/pick/008_pudding_box/config.yml b/rl/prepare_groceries/pick/008_pudding_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..f3126605c90f8b25889ac7f493369a8d955d5fb2 --- /dev/null +++ b/rl/prepare_groceries/pick/008_pudding_box/config.yml @@ -0,0 +1,113 @@ +seed: 5227 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-008_pudding_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/008_pudding_box/policy.pt" diff --git a/rl/prepare_groceries/pick/008_pudding_box/policy.pt b/rl/prepare_groceries/pick/008_pudding_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..102df87d3e1222da44d16b8f470cace6096da6fe --- /dev/null +++ b/rl/prepare_groceries/pick/008_pudding_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:092289867d462c7818abc0d452207edb305cf12c4be251b26eba6396377ffdc3 +size 55487432 diff --git a/rl/prepare_groceries/pick/009_gelatin_box/config.yml b/rl/prepare_groceries/pick/009_gelatin_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..fa2bc6b71a682017dba0a1a16920cb12ce67b359 --- /dev/null +++ b/rl/prepare_groceries/pick/009_gelatin_box/config.yml @@ -0,0 +1,113 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-009_gelatin_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/009_gelatin_box/policy.pt" diff --git a/rl/prepare_groceries/pick/009_gelatin_box/policy.pt b/rl/prepare_groceries/pick/009_gelatin_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..e6397418f688ca50742cbc9066fbc4d1136a3a27 --- /dev/null +++ b/rl/prepare_groceries/pick/009_gelatin_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:832117171b3829047b6f2f661abd769bb17ef6c47fb580572478d82f8f034ee3 +size 55487432 diff --git a/rl/prepare_groceries/pick/010_potted_meat_can/config.yml b/rl/prepare_groceries/pick/010_potted_meat_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..b5b004ff79dd8786cea4852ff9cbe1284d47b6ce --- /dev/null +++ b/rl/prepare_groceries/pick/010_potted_meat_can/config.yml @@ -0,0 +1,110 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-010_potted_meat_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/010_potted_meat_can/policy.pt" diff --git a/rl/prepare_groceries/pick/010_potted_meat_can/policy.pt b/rl/prepare_groceries/pick/010_potted_meat_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..850139b4803f4d3288653492eb51e313e5bbd8a3 --- /dev/null +++ b/rl/prepare_groceries/pick/010_potted_meat_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a04f8e0243b9ba59f50e2ae07d9269f34334d95da45e88897f1cd10d6dab5824 +size 55487432 diff --git a/rl/prepare_groceries/pick/024_bowl/config.yml b/rl/prepare_groceries/pick/024_bowl/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..4e5f46d7ba694ee24714c09bc8bffb107beedcf4 --- /dev/null +++ b/rl/prepare_groceries/pick/024_bowl/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-024_bowl + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/024_bowl/policy.pt" diff --git a/rl/prepare_groceries/pick/024_bowl/policy.pt b/rl/prepare_groceries/pick/024_bowl/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..e32fdb8d25a9606921df6e0664fb229f5259ea21 --- /dev/null +++ b/rl/prepare_groceries/pick/024_bowl/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:814686bd77abb5defd041b3ebf8edbdce26701255125843601ab03430e0d8180 +size 55487706 diff --git a/rl/prepare_groceries/pick/all/config.yml b/rl/prepare_groceries/pick/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..1761e9764eadc9d54e4efb110cf76f6a2d6a18e2 --- /dev/null +++ b/rl/prepare_groceries/pick/all/config.yml @@ -0,0 +1,113 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/pick/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-pick-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/pick/all/policy.pt" diff --git a/rl/prepare_groceries/pick/all/policy.pt b/rl/prepare_groceries/pick/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..9f26c585c0961e50d9bddb10eb229f38c64ee6a9 --- /dev/null +++ b/rl/prepare_groceries/pick/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29ae12a7bbac520ae4b0c46abfa955da294b6a09e5467abe6ecfbf9d672de52b +size 55487432 diff --git a/rl/prepare_groceries/place/002_master_chef_can/config.yml b/rl/prepare_groceries/place/002_master_chef_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..846b496d2a07393e93b3578189270cdc37df93e3 --- /dev/null +++ b/rl/prepare_groceries/place/002_master_chef_can/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-002_master_chef_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/002_master_chef_can/policy.pt" diff --git a/rl/prepare_groceries/place/002_master_chef_can/policy.pt b/rl/prepare_groceries/place/002_master_chef_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..8d4adb3adefd189867fc66f87b6cee33fad64aa5 --- /dev/null +++ b/rl/prepare_groceries/place/002_master_chef_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab001cd4b88ee6cf115c6c65a67c40b9953b347610540b17b3d524e3b82f7bde +size 55487706 diff --git a/rl/prepare_groceries/place/003_cracker_box/config.yml b/rl/prepare_groceries/place/003_cracker_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..b50ed0969411224b5cfb1408e0b0c148b8b64b5d --- /dev/null +++ b/rl/prepare_groceries/place/003_cracker_box/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-003_cracker_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/003_cracker_box/policy.pt" diff --git a/rl/prepare_groceries/place/003_cracker_box/policy.pt b/rl/prepare_groceries/place/003_cracker_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..80af2c55e37683c1fe3b51ea51536e164d003f70 --- /dev/null +++ b/rl/prepare_groceries/place/003_cracker_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc5a60777bb9cbbc3287de2b66e908913ffde5ed2c289096bd06702cbcf6f09a +size 55487706 diff --git a/rl/prepare_groceries/place/004_sugar_box/config.yml b/rl/prepare_groceries/place/004_sugar_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..ceab525776e29bb6a0a8d38a6cd0d7a70bcaa24c --- /dev/null +++ b/rl/prepare_groceries/place/004_sugar_box/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-004_sugar_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/004_sugar_box/policy.pt" diff --git a/rl/prepare_groceries/place/004_sugar_box/policy.pt b/rl/prepare_groceries/place/004_sugar_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..1c5a3edceead20a90b23ca5aac20d945e4a4493e --- /dev/null +++ b/rl/prepare_groceries/place/004_sugar_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b9cace25d850c05712488fd4cbeb409c08c2f73cd0fcfa253e411b434a7a1c2 +size 55487706 diff --git a/rl/prepare_groceries/place/005_tomato_soup_can/config.yml b/rl/prepare_groceries/place/005_tomato_soup_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..fbdff20f6838acf8c88d6e01d5bdd6b308f13446 --- /dev/null +++ b/rl/prepare_groceries/place/005_tomato_soup_can/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-005_tomato_soup_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/005_tomato_soup_can/policy.pt" diff --git a/rl/prepare_groceries/place/005_tomato_soup_can/policy.pt b/rl/prepare_groceries/place/005_tomato_soup_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..e4bd9fa2dcfa7b5b6da6d79dbf49342ed754ac3b --- /dev/null +++ b/rl/prepare_groceries/place/005_tomato_soup_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbe87309a420c505231a349b10c2d27e25e5cf481f3b485f301e7267244cebfe +size 55487706 diff --git a/rl/prepare_groceries/place/007_tuna_fish_can/config.yml b/rl/prepare_groceries/place/007_tuna_fish_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..bc88cdf15fc1926848e3a4abfa18015d6e9d3f78 --- /dev/null +++ b/rl/prepare_groceries/place/007_tuna_fish_can/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-007_tuna_fish_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/007_tuna_fish_can/policy.pt" diff --git a/rl/prepare_groceries/place/007_tuna_fish_can/policy.pt b/rl/prepare_groceries/place/007_tuna_fish_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..f2c74589662d45b90376d594292c4a0fb035be3a --- /dev/null +++ b/rl/prepare_groceries/place/007_tuna_fish_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a142a4f53b76fa20c185c2b44bb5396f40755d83005b5f2b1a4b3dab25da5d1c +size 55487706 diff --git a/rl/prepare_groceries/place/008_pudding_box/config.yml b/rl/prepare_groceries/place/008_pudding_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..254a76d696e2b0bda0cc50c0af29f7b1a5f12ea9 --- /dev/null +++ b/rl/prepare_groceries/place/008_pudding_box/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-008_pudding_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/008_pudding_box/policy.pt" diff --git a/rl/prepare_groceries/place/008_pudding_box/policy.pt b/rl/prepare_groceries/place/008_pudding_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..68a3e3e56ddcce8c86500fbf4508c9b363edba59 --- /dev/null +++ b/rl/prepare_groceries/place/008_pudding_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74cd52161c6ba6ac453dc61718265eb18a63b0d907bd383aec5ce2338ea6c185 +size 55487706 diff --git a/rl/prepare_groceries/place/009_gelatin_box/config.yml b/rl/prepare_groceries/place/009_gelatin_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..c1c38b552f3c8dbb9b000feb536b6c5cff48a70d --- /dev/null +++ b/rl/prepare_groceries/place/009_gelatin_box/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-009_gelatin_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/009_gelatin_box/policy.pt" diff --git a/rl/prepare_groceries/place/009_gelatin_box/policy.pt b/rl/prepare_groceries/place/009_gelatin_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..4f272a2d9c10b727158c3f732e84fdc114571e6a --- /dev/null +++ b/rl/prepare_groceries/place/009_gelatin_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43cac97fb30f00bf197afbb041101787addad41a62ddf165f5b6d7d309d52c99 +size 55487706 diff --git a/rl/prepare_groceries/place/010_potted_meat_can/config.yml b/rl/prepare_groceries/place/010_potted_meat_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..78b13c164b34847fd37ed4f316504de9d2b3cb54 --- /dev/null +++ b/rl/prepare_groceries/place/010_potted_meat_can/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-010_potted_meat_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/010_potted_meat_can/policy.pt" diff --git a/rl/prepare_groceries/place/010_potted_meat_can/policy.pt b/rl/prepare_groceries/place/010_potted_meat_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..8b57271710b0064ef4c673aef701891fde82d1b5 --- /dev/null +++ b/rl/prepare_groceries/place/010_potted_meat_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f15ac8f0f6e8ae41cd8332ec48cc570d254a737d377e4ce1038d09e3adea6e05 +size 55487706 diff --git a/rl/prepare_groceries/place/024_bowl/config.yml b/rl/prepare_groceries/place/024_bowl/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..c04fa680bc3c03f51c08a91d40819377e24082ff --- /dev/null +++ b/rl/prepare_groceries/place/024_bowl/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-024_bowl + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/024_bowl/policy.pt" diff --git a/rl/prepare_groceries/place/024_bowl/policy.pt b/rl/prepare_groceries/place/024_bowl/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..db08f3127efa71033ae8eddc27053ce0c5cebe7a --- /dev/null +++ b/rl/prepare_groceries/place/024_bowl/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbe6b747432c712bfbfcd65501e665ee494266b67a1bbc6cccdb35173062ef17 +size 55487706 diff --git a/rl/prepare_groceries/place/all/config.yml b/rl/prepare_groceries/place/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..24ed95dd84e2c9935e21c1cad880dc6eac466d91 --- /dev/null +++ b/rl/prepare_groceries/place/all/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/prepare_groceries/place/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/prepare_groceries/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-prepare_groceries-place-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/prepare_groceries/place/all/policy.pt" diff --git a/rl/prepare_groceries/place/all/policy.pt b/rl/prepare_groceries/place/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..4bf0bb6a85eb12ea41cc66132a5597ce34114cd5 --- /dev/null +++ b/rl/prepare_groceries/place/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd883cd7e98a994e5cce57c15549c6e03e0ee0052cc3494a4f03d43ad106bfae +size 55487706 diff --git a/rl/set_table/close/fridge/config.yml b/rl/set_table/close/fridge/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..c88f170705771f3b6303eba6151a451a1e04ac3c --- /dev/null +++ b/rl/set_table/close/fridge/config.yml @@ -0,0 +1,85 @@ +seed: 2337 +env: + env_id: CloseSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/close/train/fridge.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/close/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: CloseSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/close/train/fridge.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/close/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 10 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: ppo + total_timesteps: 25000000 + learning_rate: 0.0003 + anneal_lr: false + gamma: 0.9 + gae_lambda: 0.9 + num_minibatches: 16 + update_epochs: 8 + norm_adv: true + clip_coef: 0.2 + clip_vloss: false + ent_coef: 0.0 + vf_coef: 0.5 + max_grad_norm: 0.5 + target_kl: 0.2 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + finite_horizon_gae: true + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_steps: 100 + num_envs: 189 + num_eval_envs: 63 + batch_size: 18900 + minibatch_size: 1181 + num_iterations: 1323 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-close-fridge + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/close/fridge/policy.pt" diff --git a/rl/set_table/close/fridge/policy.pt b/rl/set_table/close/fridge/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..da8b7e30cfdc5adb882e20bc09e0bcb1e5fef22a --- /dev/null +++ b/rl/set_table/close/fridge/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:927bfb486c1eaf0ed8772fd7300be3ebf49d7bf8f18f24d141178c77fc4814ad +size 68155930 diff --git a/rl/set_table/close/kitchen_counter/config.yml b/rl/set_table/close/kitchen_counter/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..4abe3013b026ea80c811b4852e2db08ab92d2412 --- /dev/null +++ b/rl/set_table/close/kitchen_counter/config.yml @@ -0,0 +1,85 @@ +seed: 4682 +env: + env_id: CloseSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/close/train/kitchen_counter.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/close/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: CloseSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/close/train/kitchen_counter.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/close/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 10 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: ppo + total_timesteps: 15000000 + learning_rate: 0.0003 + anneal_lr: false + gamma: 0.9 + gae_lambda: 0.9 + num_minibatches: 16 + update_epochs: 8 + norm_adv: true + clip_coef: 0.2 + clip_vloss: false + ent_coef: 0.0 + vf_coef: 0.5 + max_grad_norm: 0.5 + target_kl: 0.2 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + finite_horizon_gae: true + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_steps: 100 + num_envs: 189 + num_eval_envs: 63 + batch_size: 18900 + minibatch_size: 1181 + num_iterations: 794 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-close-kitchen_counter + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/close/kitchen_counter/policy.pt" diff --git a/rl/set_table/close/kitchen_counter/policy.pt b/rl/set_table/close/kitchen_counter/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..092ec56f53cbe83a3a2b422f0527c3c22a31baed --- /dev/null +++ b/rl/set_table/close/kitchen_counter/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23b2552d7e077ce8c8e1d521a4fad520aa7e82c027521c2458c8d5c6938ebd92 +size 68155930 diff --git a/rl/set_table/open/fridge/config.yml b/rl/set_table/open/fridge/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..2928641101726975188eda18777b939ed8b241fa --- /dev/null +++ b/rl/set_table/open/fridge/config.yml @@ -0,0 +1,87 @@ +seed: 2337 +env: + env_id: OpenSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/open/train/fridge.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/open/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false + randomly_slightly_open_articulation: false +eval_env: + env_id: OpenSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/open/train/fridge.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/open/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 10 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false + randomly_slightly_open_articulation: false +algo: + name: ppo + total_timesteps: 15000000 + learning_rate: 0.0003 + anneal_lr: false + gamma: 0.9 + gae_lambda: 0.9 + num_minibatches: 16 + update_epochs: 8 + norm_adv: true + clip_coef: 0.2 + clip_vloss: false + ent_coef: 0.0 + vf_coef: 0.5 + max_grad_norm: 0.5 + target_kl: 0.2 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + finite_horizon_gae: true + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_steps: 100 + num_envs: 189 + num_eval_envs: 63 + batch_size: 18900 + minibatch_size: 1181 + num_iterations: 794 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-open-fridge + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/open/fridge/policy.pt" diff --git a/rl/set_table/open/fridge/policy.pt b/rl/set_table/open/fridge/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..70bcda8f61e0ab0ff282121f1373c35477232bd2 --- /dev/null +++ b/rl/set_table/open/fridge/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d863fa8642584ccdfe2834a6a9b3baa34884d399a5e221fd682639fccdcc879a +size 68155930 diff --git a/rl/set_table/open/kitchen_counter/config.yml b/rl/set_table/open/kitchen_counter/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..b90fbd260c7baa99f00b016e6c87a17b5f597550 --- /dev/null +++ b/rl/set_table/open/kitchen_counter/config.yml @@ -0,0 +1,87 @@ +seed: 4682 +env: + env_id: OpenSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/open/train/kitchen_counter.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/open/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false + randomly_slightly_open_articulation: true +eval_env: + env_id: OpenSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/open/train/kitchen_counter.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/open/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: 10 + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false + randomly_slightly_open_articulation: false +algo: + name: ppo + total_timesteps: 50000000 + learning_rate: 0.0003 + anneal_lr: false + gamma: 0.9 + gae_lambda: 0.9 + num_minibatches: 16 + update_epochs: 8 + norm_adv: true + clip_coef: 0.2 + clip_vloss: false + ent_coef: 0.0 + vf_coef: 0.5 + max_grad_norm: 0.5 + target_kl: 0.2 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + finite_horizon_gae: true + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_steps: 100 + num_envs: 189 + num_eval_envs: 63 + batch_size: 18900 + minibatch_size: 1181 + num_iterations: 2646 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-open-kitchen_counter + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/open/kitchen_counter/policy.pt" diff --git a/rl/set_table/open/kitchen_counter/policy.pt b/rl/set_table/open/kitchen_counter/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..be8b3b3245c324c899d005d7525f192e3690461a --- /dev/null +++ b/rl/set_table/open/kitchen_counter/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:290b22d3c23d12927b6a906e2e54670c9d90605b3e04ef7717c4dd2ad3787de1 +size 68155930 diff --git a/rl/set_table/pick/013_apple/config.yml b/rl/set_table/pick/013_apple/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..34d5582ec43a9ed8faa8a70ecc1df989555610af --- /dev/null +++ b/rl/set_table/pick/013_apple/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/013_apple.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/013_apple.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-pick-013_apple + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/pick/013_apple/policy.pt" diff --git a/rl/set_table/pick/013_apple/policy.pt b/rl/set_table/pick/013_apple/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..07ebe69fa32f6f4a0e5b1de2bd9606cf28000969 --- /dev/null +++ b/rl/set_table/pick/013_apple/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d77da6845a190768792d08a63c2f3e6b19f1022865c350b9ea34f0c9b3aa5cb +size 55487706 diff --git a/rl/set_table/pick/024_bowl/config.yml b/rl/set_table/pick/024_bowl/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..16d11dc49851d2b91e11fd70875b019ea1a6bb72 --- /dev/null +++ b/rl/set_table/pick/024_bowl/config.yml @@ -0,0 +1,115 @@ +seed: 5227 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-pick-024_bowl + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/pick/024_bowl/policy.pt" diff --git a/rl/set_table/pick/024_bowl/policy.pt b/rl/set_table/pick/024_bowl/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..cdeb9dfba9aec9f735c1474463d6d03c52da1ee3 --- /dev/null +++ b/rl/set_table/pick/024_bowl/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59e44d94e78a95312fb1c64024df9258f50c508b160b697aae9a124be0796f0c +size 55487706 diff --git a/rl/set_table/pick/all/config.yml b/rl/set_table/pick/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..2b35a1c93dcdede0cfc4febe452f09c5d9333bea --- /dev/null +++ b/rl/set_table/pick/all/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/pick/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 200000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-pick-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/pick/all/policy.pt" diff --git a/rl/set_table/pick/all/policy.pt b/rl/set_table/pick/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..3301eedbe2192293eeaf79afe55b4be5a88a5d74 --- /dev/null +++ b/rl/set_table/pick/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:588ba4ad3452c38253b432de4275675f8fa41fb1d0011d29c05c560229b59e5e +size 55488538 diff --git a/rl/set_table/place/013_apple/config.yml b/rl/set_table/place/013_apple/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..1c17aaf55dfc6bfe4c807a2461c488b627c45ed2 --- /dev/null +++ b/rl/set_table/place/013_apple/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/013_apple.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/013_apple.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-place-013_apple + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/place/013_apple/policy.pt" diff --git a/rl/set_table/place/013_apple/policy.pt b/rl/set_table/place/013_apple/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..365e59769bff97a1ef5788baaca4bba114fa6884 --- /dev/null +++ b/rl/set_table/place/013_apple/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:546e9413a5d12602bcd51b320a243b9a3ec40d66cd4067630ba1f6eba877ec0e +size 55487706 diff --git a/rl/set_table/place/024_bowl/config.yml b/rl/set_table/place/024_bowl/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..7d8e2c26745c8f5b1211b6d0fcbc9de395cb29e5 --- /dev/null +++ b/rl/set_table/place/024_bowl/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-place-024_bowl + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/place/024_bowl/policy.pt" diff --git a/rl/set_table/place/024_bowl/policy.pt b/rl/set_table/place/024_bowl/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..0cc3b60ad273980a81ada7b4854c2c14e3904c80 --- /dev/null +++ b/rl/set_table/place/024_bowl/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9fdc8e66fcac84406f7d5d26414b8c920788e2a1b9243cc4b0c4835bb833e949 +size 55487706 diff --git a/rl/set_table/place/all/config.yml b/rl/set_table/place/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..796f080f8cd0c0baa2e6157fa7074378fedbe3e0 --- /dev/null +++ b/rl/set_table/place/all/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/set_table/place/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/set_table/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-set_table-place-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/set_table/place/all/policy.pt" diff --git a/rl/set_table/place/all/policy.pt b/rl/set_table/place/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..b1665d9cd6fd72bbf613d695f6cd2f11ddc3bfb1 --- /dev/null +++ b/rl/set_table/place/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ac84a91d03e158b30279997356ad9e28e6af4cdfa8a1a0423dc0930fcbcf2fa +size 55487706 diff --git a/rl/tidy_house/pick/002_master_chef_can/config.yml b/rl/tidy_house/pick/002_master_chef_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..45d75661d7d0ec04f2d72355d258962d494dbe9c --- /dev/null +++ b/rl/tidy_house/pick/002_master_chef_can/config.yml @@ -0,0 +1,113 @@ +seed: 5227 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-002_master_chef_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/002_master_chef_can/policy.pt" diff --git a/rl/tidy_house/pick/002_master_chef_can/policy.pt b/rl/tidy_house/pick/002_master_chef_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..80b5fd3827e8fbb931f1e3b88c0d456381631c23 --- /dev/null +++ b/rl/tidy_house/pick/002_master_chef_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af4a679414747056a42ff80b657b6764296e297ee91055c012ea370fdaeea54f +size 55487432 diff --git a/rl/tidy_house/pick/003_cracker_box/config.yml b/rl/tidy_house/pick/003_cracker_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..9722d38794fec7c58231c852134ffa9aa6e135ab --- /dev/null +++ b/rl/tidy_house/pick/003_cracker_box/config.yml @@ -0,0 +1,111 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-003_cracker_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/003_cracker_box/policy.pt" diff --git a/rl/tidy_house/pick/003_cracker_box/policy.pt b/rl/tidy_house/pick/003_cracker_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..b453799ae6ab73236d96b75bc38231c4b7e8b27d --- /dev/null +++ b/rl/tidy_house/pick/003_cracker_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79c5f186c6f6c37b86174189f83361a36c395f2f3d52bc80ce46dcfa6a3bd55a +size 55487432 diff --git a/rl/tidy_house/pick/004_sugar_box/config.yml b/rl/tidy_house/pick/004_sugar_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..08b03a51b7f5a1bcb75826a232a3ee721e0ef135 --- /dev/null +++ b/rl/tidy_house/pick/004_sugar_box/config.yml @@ -0,0 +1,110 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-004_sugar_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/004_sugar_box/policy.pt" diff --git a/rl/tidy_house/pick/004_sugar_box/policy.pt b/rl/tidy_house/pick/004_sugar_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..ae6aa75e3a15d330e38175d6b6ee59de82f9ec45 --- /dev/null +++ b/rl/tidy_house/pick/004_sugar_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7c9a24803aca8f0989c0271621d37bc213d73cad615304169bf3d8557b87a8a +size 55487432 diff --git a/rl/tidy_house/pick/005_tomato_soup_can/config.yml b/rl/tidy_house/pick/005_tomato_soup_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..2e3fdd816797f821beb0e658860e86512822164f --- /dev/null +++ b/rl/tidy_house/pick/005_tomato_soup_can/config.yml @@ -0,0 +1,115 @@ +seed: 5227 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-005_tomato_soup_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/005_tomato_soup_can/policy.pt" diff --git a/rl/tidy_house/pick/005_tomato_soup_can/policy.pt b/rl/tidy_house/pick/005_tomato_soup_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..ab67d97d390cd711680306ca14912d18a6ca62e5 --- /dev/null +++ b/rl/tidy_house/pick/005_tomato_soup_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fd781b0f862d297d0cbadcbc38f5285f26a36f77eb35537453ffeabcbbd6fa6 +size 55487706 diff --git a/rl/tidy_house/pick/007_tuna_fish_can/config.yml b/rl/tidy_house/pick/007_tuna_fish_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..a1b7832321ae5a796d466e9ef95b8eb1eee79f5b --- /dev/null +++ b/rl/tidy_house/pick/007_tuna_fish_can/config.yml @@ -0,0 +1,113 @@ +seed: 5227 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-007_tuna_fish_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/007_tuna_fish_can/policy.pt" diff --git a/rl/tidy_house/pick/007_tuna_fish_can/policy.pt b/rl/tidy_house/pick/007_tuna_fish_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..b62263f2ecd131fd7bedab8727bfd86d6b2fb9a2 --- /dev/null +++ b/rl/tidy_house/pick/007_tuna_fish_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26da30b9bc513341981f7999249e2ed2ffa62a0132b4029be858ceeccbff16d9 +size 55487432 diff --git a/rl/tidy_house/pick/008_pudding_box/config.yml b/rl/tidy_house/pick/008_pudding_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..839c6df2b45a5774ed12df989229ff4264274c42 --- /dev/null +++ b/rl/tidy_house/pick/008_pudding_box/config.yml @@ -0,0 +1,111 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-008_pudding_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/008_pudding_box/policy.pt" diff --git a/rl/tidy_house/pick/008_pudding_box/policy.pt b/rl/tidy_house/pick/008_pudding_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..c11ddd528ba8ad4bbff4e6853783545155f4a49b --- /dev/null +++ b/rl/tidy_house/pick/008_pudding_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:014d49de2e484e9026bf2506abac8f992ed236b95c175706bc0920c239360be1 +size 55487432 diff --git a/rl/tidy_house/pick/009_gelatin_box/config.yml b/rl/tidy_house/pick/009_gelatin_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..8aec356fe096ea89e5ee07141254333c525d1845 --- /dev/null +++ b/rl/tidy_house/pick/009_gelatin_box/config.yml @@ -0,0 +1,111 @@ +seed: 4682 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-009_gelatin_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/009_gelatin_box/policy.pt" diff --git a/rl/tidy_house/pick/009_gelatin_box/policy.pt b/rl/tidy_house/pick/009_gelatin_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..00228b9db5fdea158b8fdeb5358ef0a047b3cbdc --- /dev/null +++ b/rl/tidy_house/pick/009_gelatin_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c192650f65a8c29e939a1cf4b92af205be80caf57943ce79033e91da5a00febe +size 55487432 diff --git a/rl/tidy_house/pick/010_potted_meat_can/config.yml b/rl/tidy_house/pick/010_potted_meat_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..d187069382a1b4ae3718737125fdb091ad949cce --- /dev/null +++ b/rl/tidy_house/pick/010_potted_meat_can/config.yml @@ -0,0 +1,110 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-010_potted_meat_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/010_potted_meat_can/policy.pt" diff --git a/rl/tidy_house/pick/010_potted_meat_can/policy.pt b/rl/tidy_house/pick/010_potted_meat_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..b117b141d7b2d688d874694a4fa7d6a83a56acc5 --- /dev/null +++ b/rl/tidy_house/pick/010_potted_meat_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31633e37cbc13fd36751f57cd32c3f2690080c144d561fc0888fc7c6572d2f03 +size 55487432 diff --git a/rl/tidy_house/pick/024_bowl/config.yml b/rl/tidy_house/pick/024_bowl/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..0f387eefa5b395c0d1842127ca52232fd1944ce5 --- /dev/null +++ b/rl/tidy_house/pick/024_bowl/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-024_bowl + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/024_bowl/policy.pt" diff --git a/rl/tidy_house/pick/024_bowl/policy.pt b/rl/tidy_house/pick/024_bowl/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..507e53d145c702ca86b82cba95568ecc78a85c5f --- /dev/null +++ b/rl/tidy_house/pick/024_bowl/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4d80a64630eb16adb302c2095189c423e8cf0bb0b929772e03fbcbe6dae5f4a +size 55487706 diff --git a/rl/tidy_house/pick/all/config.yml b/rl/tidy_house/pick/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..b2f00c687b4bc02043a9d75803187c3eefddc0d1 --- /dev/null +++ b/rl/tidy_house/pick/all/config.yml @@ -0,0 +1,113 @@ +seed: 2337 +env: + env_id: PickSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PickSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/pick/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/pick/train/spawn_data.pt + record_video: false + debug_video: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-pick-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/pick/all/policy.pt" diff --git a/rl/tidy_house/pick/all/policy.pt b/rl/tidy_house/pick/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..6e08b3116eecc7d73ff0e95996878f4f3c4e4702 --- /dev/null +++ b/rl/tidy_house/pick/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81f1f29b3af1ed4a3d3b502b515595dae852c4d842a1b1599203e46dee5f73e9 +size 55487432 diff --git a/rl/tidy_house/place/002_master_chef_can/config.yml b/rl/tidy_house/place/002_master_chef_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..1b9681d0479f0d14d7fc1a711b0c323cb6bae9de --- /dev/null +++ b/rl/tidy_house/place/002_master_chef_can/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/002_master_chef_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-002_master_chef_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/002_master_chef_can/policy.pt" diff --git a/rl/tidy_house/place/002_master_chef_can/policy.pt b/rl/tidy_house/place/002_master_chef_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..32f3869ccdbe1b920892acc29b9daf12af4ffb2f --- /dev/null +++ b/rl/tidy_house/place/002_master_chef_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:155db70f9b84f54942f5df5c7ab740f24efb9dc5aef7b0091926b2521c9c59f8 +size 55487706 diff --git a/rl/tidy_house/place/003_cracker_box/config.yml b/rl/tidy_house/place/003_cracker_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..bf3adb93005786682f2aca8e7c3fafc8e941ab83 --- /dev/null +++ b/rl/tidy_house/place/003_cracker_box/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/003_cracker_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-003_cracker_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/003_cracker_box/policy.pt" diff --git a/rl/tidy_house/place/003_cracker_box/policy.pt b/rl/tidy_house/place/003_cracker_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..d2b385eb3834975231cdaee91e0f938c79f302fa --- /dev/null +++ b/rl/tidy_house/place/003_cracker_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b3cd190ff1f18131c8c7e2ffb934e0f864622876e8019f1f46b291783d23d5b +size 55487706 diff --git a/rl/tidy_house/place/004_sugar_box/config.yml b/rl/tidy_house/place/004_sugar_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..e7d1503d90b6f924b5c57cc3a39fe9406c3a7b02 --- /dev/null +++ b/rl/tidy_house/place/004_sugar_box/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/004_sugar_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-004_sugar_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/004_sugar_box/policy.pt" diff --git a/rl/tidy_house/place/004_sugar_box/policy.pt b/rl/tidy_house/place/004_sugar_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..a66a8e83a2e193f849b22323d03a15acb7c06241 --- /dev/null +++ b/rl/tidy_house/place/004_sugar_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4e4c04935bf7d21f1c00107cbab1715856385338309749e24827406b52811c3 +size 55487706 diff --git a/rl/tidy_house/place/005_tomato_soup_can/config.yml b/rl/tidy_house/place/005_tomato_soup_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..6493a9e6134f923a95a3d35cac2477a722b5ac10 --- /dev/null +++ b/rl/tidy_house/place/005_tomato_soup_can/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/005_tomato_soup_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-005_tomato_soup_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/005_tomato_soup_can/policy.pt" diff --git a/rl/tidy_house/place/005_tomato_soup_can/policy.pt b/rl/tidy_house/place/005_tomato_soup_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..696a3021c734ddf81452a85ca53ae30c774e460e --- /dev/null +++ b/rl/tidy_house/place/005_tomato_soup_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5d555039b617fe316aba297705085895fc1b2e975676b51e7efde9fe481ce53 +size 55487706 diff --git a/rl/tidy_house/place/007_tuna_fish_can/config.yml b/rl/tidy_house/place/007_tuna_fish_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..b7162c27286468bfeac9fc010dc47de4c6639824 --- /dev/null +++ b/rl/tidy_house/place/007_tuna_fish_can/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/007_tuna_fish_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-007_tuna_fish_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/007_tuna_fish_can/policy.pt" diff --git a/rl/tidy_house/place/007_tuna_fish_can/policy.pt b/rl/tidy_house/place/007_tuna_fish_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..48403246270f30e576472e2fc7a7ccc6771e9f64 --- /dev/null +++ b/rl/tidy_house/place/007_tuna_fish_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3333fff1de5e1c42112a85fb594330606350735ebeb358877321b2f1db3a654e +size 55487706 diff --git a/rl/tidy_house/place/008_pudding_box/config.yml b/rl/tidy_house/place/008_pudding_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..128057b93b82e26b2a38dbfea0759f9ca0985de2 --- /dev/null +++ b/rl/tidy_house/place/008_pudding_box/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/008_pudding_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-008_pudding_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/008_pudding_box/policy.pt" diff --git a/rl/tidy_house/place/008_pudding_box/policy.pt b/rl/tidy_house/place/008_pudding_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..af0157159d273d1b7004fe14dfe53f6e6f1db057 --- /dev/null +++ b/rl/tidy_house/place/008_pudding_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b24d4757f231b3ef6661f7230b7d44cbd6f789da2a01a73b8b512a58e3df6f78 +size 55487706 diff --git a/rl/tidy_house/place/009_gelatin_box/config.yml b/rl/tidy_house/place/009_gelatin_box/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..715073576b23a9c11c5a45f21ee9419af1f66800 --- /dev/null +++ b/rl/tidy_house/place/009_gelatin_box/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/009_gelatin_box.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-009_gelatin_box + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/009_gelatin_box/policy.pt" diff --git a/rl/tidy_house/place/009_gelatin_box/policy.pt b/rl/tidy_house/place/009_gelatin_box/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..3b4416c5504d7cd0a97b48f196da5aa450e0b7a9 --- /dev/null +++ b/rl/tidy_house/place/009_gelatin_box/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3bdb97f6dcc51d0301f72fee91aebfc0218051a8f55820dc53889d801bddb88e +size 55487706 diff --git a/rl/tidy_house/place/010_potted_meat_can/config.yml b/rl/tidy_house/place/010_potted_meat_can/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..683ef88d08eaeff169f551e1292c3875d6d4bdaf --- /dev/null +++ b/rl/tidy_house/place/010_potted_meat_can/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/010_potted_meat_can.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-010_potted_meat_can + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/010_potted_meat_can/policy.pt" diff --git a/rl/tidy_house/place/010_potted_meat_can/policy.pt b/rl/tidy_house/place/010_potted_meat_can/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..7b20ecca030e2838d97617111d75df42bdb9d18f --- /dev/null +++ b/rl/tidy_house/place/010_potted_meat_can/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b636e6cf0e3d4dc8ae2724a5f6efe342f9ceddbb2aa11f680340071d723d104d +size 55487706 diff --git a/rl/tidy_house/place/024_bowl/config.yml b/rl/tidy_house/place/024_bowl/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..de3f9d33d3283261a5e981e892be1cadce3af8c1 --- /dev/null +++ b/rl/tidy_house/place/024_bowl/config.yml @@ -0,0 +1,115 @@ +seed: 4682 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 189 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/024_bowl.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 25000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 189 + num_iterations: 396826 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-024_bowl + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/024_bowl/policy.pt" diff --git a/rl/tidy_house/place/024_bowl/policy.pt b/rl/tidy_house/place/024_bowl/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..ed24fc55753968d45da90d5bfac1644008b3d56f --- /dev/null +++ b/rl/tidy_house/place/024_bowl/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff82d989e5e1e7d549574ef8037420721c57ffc01723e86e63d904b93799b942 +size 55487706 diff --git a/rl/tidy_house/place/all/config.yml b/rl/tidy_house/place/all/config.yml new file mode 100644 index 0000000000000000000000000000000000000000..7bf49c77b265e2ee4db887150c8c98b06c6d9c71 --- /dev/null +++ b/rl/tidy_house/place/all/config.yml @@ -0,0 +1,115 @@ +seed: 2337 +env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 100 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: false + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +eval_env: + env_id: PlaceSubtaskTrain-v0 + num_envs: 63 + max_episode_steps: 200 + make_env: true + continuous_task: true + cat_state: true + cat_pixels: false + frame_stack: 3 + stationary_base: false + stationary_torso: false + stationary_head: true + task_plan_fp: task_plans/tidy_house/place/train/all.json + spawn_data_fp: /arshukla-fast-vol-1/ManiSkill/data/scene_datasets/replica_cad_dataset/rearrange/spawn_data/tidy_house/place/train/spawn_data.pt + record_video: false + debug_video: false + debug_video_gen: false + save_video_freq: null + info_on_video: true + extra_stat_keys: [] + env_kwargs: + robot_force_mult: 0.001 + robot_force_penalty_min: 0.2 + target_randomization: false +algo: + name: sac + replay_buffer_capacity: 995400 + total_timesteps: 50000000 + num_steps: 100 + init_steps: 5000 + batch_size: 512 + critic_encoder_tau: 0.005 + cnn_features: + - 32 + - 64 + - 128 + - 256 + cnn_filters: + - 3 + - 3 + - 3 + - 3 + cnn_strides: + - 2 + - 2 + - 2 + - 2 + cnn_padding: valid + encoder_pixels_feature_dim: 50 + encoder_state_feature_dim: 50 + detach_encoder: false + critic_hidden_dims: + - 256 + - 256 + - 256 + critic_lr: 0.0003 + critic_layer_norm: true + critic_dropout: null + critic_beta: 0.9 + critic_tau: 0.005 + critic_target_update_freq: 2 + actor_hidden_dims: + - 256 + - 256 + - 256 + actor_lr: 0.0003 + actor_beta: 0.9 + actor_log_std_min: -20 + actor_log_std_max: 2 + actor_update_freq: 2 + gamma: 0.9 + init_temperature: 0.1 + alpha_lr: 0.0003 + alpha_beta: 0.9 + log_freq: 10000 + save_freq: 100000 + eval_freq: 100000 + torch_deterministic: true + save_backup_ckpts: false + eval_episodes: 189 + num_envs: 63 + num_eval_envs: 63 + num_iterations: 793651 +logger: + workspace: mshab_exps + exp_name: rcad-tidy_house-place-all + clear_out: true + tensorboard: true + wandb: false +model_ckpt: "mshab_checkpoints/rl/tidy_house/place/all/policy.pt" diff --git a/rl/tidy_house/place/all/policy.pt b/rl/tidy_house/place/all/policy.pt new file mode 100644 index 0000000000000000000000000000000000000000..dcd1ead0ea5544ba2a568119a4ee4a75f7558ee4 --- /dev/null +++ b/rl/tidy_house/place/all/policy.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9c1b9b27c692c2dcd313db818ac9d74a0cef01ca9aa37de630e2ebf18ae556b +size 55487706