bc and rl checkpoints

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"

bc/prepare_groceries/pick/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b90849268bcaefe1fa1c1abd79f9e259a48f42617bc152bdeb8943dbd39a45c
+size 238201234

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"

bc/prepare_groceries/place/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4bc570e9aad369d26c62d69c5e441f7a46fc45ec20221c5c6f3d11708a49f6e
+size 238201234

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"

bc/set_table/close/fridge/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4e6802c0b767f70b819f74eaf1b063d028e0592cf6ad0cf47739439bf2f1a3f
+size 238200850

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"

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

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"

bc/set_table/open/fridge/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80f614cdfbef42079f9f4254224484d71d49057801182417628d181646cefe38
+size 238200850

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"

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

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"

bc/set_table/pick/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3593b72dfc3c265fb8a294b6f3bb79c86472d4efada6437d50c66daee7cbd58
+size 238200850

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"

bc/set_table/place/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39cb5b0a2000c8184a5e4a25cda28e60602734a81e05be972a18f27f911bda2a
+size 238200850

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"

bc/tidy_house/pick/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:250f92d9a0d6c42f78ba85ac0cab740eee1f80619f6f6fbf39b7ad8aa676d3c8
+size 238201234

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"

bc/tidy_house/place/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e947b7ab786abd6817536bd7af53897f97c89d1214d06ac0c025df5c915b0f2
+size 238200850

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"

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

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"

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

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"

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

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"

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

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"

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

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"

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

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"

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

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"

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

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"

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

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"

rl/prepare_groceries/pick/all/policy.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29ae12a7bbac520ae4b0c46abfa955da294b6a09e5467abe6ecfbf9d672de52b
+size 55487432

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"

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

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"

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

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"

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

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"

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

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"

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