diff --git a/N053/go2_arm/dr_on/seed_2119/rec/check.json b/N053/go2_arm/dr_on/seed_2119/rec/check.json index b183204e65dbb2ec240278c3d9fb6dbcf40326d5..31adb2db76d5f0459c883735f6d3deef71e02069 100644 --- a/N053/go2_arm/dr_on/seed_2119/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2119/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2119/rec/check.txt b/N053/go2_arm/dr_on/seed_2119/rec/check.txt index 6319e907473c96a45c9eb6d74fc4fe7e985b3b0b..3cac4365c8421ce457c6265a3bfc73821416c993 100644 --- a/N053/go2_arm/dr_on/seed_2119/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2119/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2119/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1315 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1315 at dataset fps 25 = 52.60 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1315 rows vs ceil(2629/2) = 1315 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2689, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 11b064c874776a45090a64803e902edb42c6e9d2..757062046fdc65e981c7bc77fac7396e57a5fec7 100644 --- a/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1315, "magicsim/control_steps": 2629, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1315, "magicsim/control_steps": 2629, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index cdb2267a4966394204e27868835eda637e88f4b7..d75b9098a17aafdade0e6cd0bca9dc22f384d174 100644 --- a/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2119/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2120/rec/check.json b/N053/go2_arm/dr_on/seed_2120/rec/check.json index 8c595f1323e1f9c14700544873afdf9729d5e5ab..f5e61b41cb1217c0d3b9441188f5bb33c5db4a8e 100644 --- a/N053/go2_arm/dr_on/seed_2120/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2120/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2120/rec/check.txt b/N053/go2_arm/dr_on/seed_2120/rec/check.txt index 0ff075366b02cef78fb5de79452033bcdfced626..59cbb4b866509bd5fb827dcee5ff0637243e03c3 100644 --- a/N053/go2_arm/dr_on/seed_2120/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2120/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2120/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1180 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1180 at dataset fps 25 = 47.20 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1180 rows vs ceil(2359/2) = 1180 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2419, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index c9f0d5135fe06e1b7b4f50a49cd1e7f5e78b6170..611f9acc46ce1782ef834b5f227063655f5fd98c 100644 --- a/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1180, "magicsim/control_steps": 2359, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1180, "magicsim/control_steps": 2359, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 2a0c40d4c3d5e70c64eb75f246870e7852bf3c31..17920714732c0670748139502fbeb37375e12601 100644 --- a/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2120/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2121/rec/check.json b/N053/go2_arm/dr_on/seed_2121/rec/check.json index 07ae934faa2fd230a9438dc15ab9fac2e46d6ebf..85a2cdfdaca60121ffff26e6c8254439a9b2bdac 100644 --- a/N053/go2_arm/dr_on/seed_2121/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2121/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2121/rec/check.txt b/N053/go2_arm/dr_on/seed_2121/rec/check.txt index def32c91ca340fdf8c1f3b7763b4f0109d5f5a9e..a5ec4c0616fb9d64e256a7d909190ff1ceafda6b 100644 --- a/N053/go2_arm/dr_on/seed_2121/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2121/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2121/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1174 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1174 at dataset fps 25 = 46.96 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1174 rows vs ceil(2347/2) = 1174 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2407, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 87f839cb389fefe01d3de0958c43153e4b7e9b94..d840a31ee752422ef9bb3d8ef23a25d10fa2b662 100644 --- a/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1174, "magicsim/control_steps": 2347, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1174, "magicsim/control_steps": 2347, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 6d5c41acecd143227822ccc24ad8f57287157c60..93811e4e1fa9bf7b90386253274f9116733c8980 100644 --- a/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2121/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2122/rec/check.json b/N053/go2_arm/dr_on/seed_2122/rec/check.json index 30a3f0a1808f4c0c860b43ebbd7b30dcaccb1b54..73f57ccd22fc7288a008998d3a6773a78a679fa3 100644 --- a/N053/go2_arm/dr_on/seed_2122/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2122/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2122/rec/check.txt b/N053/go2_arm/dr_on/seed_2122/rec/check.txt index df010aea919ae1c7b38b90eb5001ca636e389ce0..0c44ec591b7edeb9e00a8cb3aa350af9b953c31f 100644 --- a/N053/go2_arm/dr_on/seed_2122/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2122/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2122_r1/out/rec/memorystage/0 --- episode file episode_000000.parquet: 1241 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1241 at dataset fps 25 = 49.64 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1241 rows vs ceil(2482/2) = 1241 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2541, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index e853b8f8f8222621c7689603ddccece0e1128e78..3d1e8aba2fbfd2ad8e083e05ade9532e0f5377e0 100644 --- a/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1241, "magicsim/control_steps": 2482, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1241, "magicsim/control_steps": 2482, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 969e2bfbb26b008e6eb6336de449902eab772c38..bd919651f698472043f57399099e7b728cdb6c3b 100644 --- a/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2122/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2123/rec/check.json b/N053/go2_arm/dr_on/seed_2123/rec/check.json index 2f32e7b70cd9b69d9ecaa8f0ddc2b8731ed5f1e1..1fa0050291d59ab35107495f1e7e3770c25e2064 100644 --- a/N053/go2_arm/dr_on/seed_2123/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2123/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2123/rec/check.txt b/N053/go2_arm/dr_on/seed_2123/rec/check.txt index c14ab37a1513fcf0a9dbf969388fa7adf7e665aa..ef0317233cc2cdd2bed1d9b0cd2fb6a9461cd28e 100644 --- a/N053/go2_arm/dr_on/seed_2123/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2123/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2123/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1172 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1172 at dataset fps 25 = 46.88 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1172 rows vs ceil(2344/2) = 1172 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2403, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 7ecb7cc3707c78188d6e1fdcc5f31b3b20085371..f6c447ec7d44f0f0c83e1fb88a6fa14124f8c6ab 100644 --- a/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1172, "magicsim/control_steps": 2344, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1172, "magicsim/control_steps": 2344, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index e5f307056283f60020b5bc430e656c61850d0cb4..74179cce13574ffd89fbdfabae0fc83195ea10bc 100644 --- a/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2123/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2124/rec/check.json b/N053/go2_arm/dr_on/seed_2124/rec/check.json index 4b926b2d4c0b6bc2a4f308ee53bd0c11fbb070e6..e888d37337c9517b87d5a2eea59d54f2ce802bba 100644 --- a/N053/go2_arm/dr_on/seed_2124/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2124/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2124/rec/check.txt b/N053/go2_arm/dr_on/seed_2124/rec/check.txt index 5b08cd886ad93463047424df3436385c986a5628..b60a3dd47338e5d2bf7ed5055d6c50e66de4bd2f 100644 --- a/N053/go2_arm/dr_on/seed_2124/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2124/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2124_r1/out/rec/memorystage/0 --- episode file episode_000000.parquet: 1248 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1248 at dataset fps 25 = 49.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1248 rows vs ceil(2495/2) = 1248 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2555, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index c7aeb818b06ce61e97ea3ca4f1830c205f2c5932..46a9716008c3f000a91d7273a8a3511c867d4677 100644 --- a/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1248, "magicsim/control_steps": 2495, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1248, "magicsim/control_steps": 2495, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 28e6efcc92c7f4872a2117fc804cbe409970fb59..458aadf78fd5b5d7ccc3d87f1c3a3807ed406a1d 100644 --- a/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2124/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2125/rec/check.json b/N053/go2_arm/dr_on/seed_2125/rec/check.json index 8ac4553c6c5ee0ec4e0431b56d99a4772367a2df..8787c477c1256ad32be0c0930be7e6cf5c270ea1 100644 --- a/N053/go2_arm/dr_on/seed_2125/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2125/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2125/rec/check.txt b/N053/go2_arm/dr_on/seed_2125/rec/check.txt index 56504f67f2fc1b5a7cf9d760d3a50f8041742c9d..62fa3e769c867816084ee12fccafcefbdedfbe14 100644 --- a/N053/go2_arm/dr_on/seed_2125/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2125/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2125/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1215 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1215 at dataset fps 25 = 48.60 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1215 rows vs ceil(2430/2) = 1215 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2489, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 485b400a3b8fcd6e6aee5eb3f54a5fd86a8922bb..4ad440c2f2f150a0a880a669df4119151daf55d8 100644 --- a/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1215, "magicsim/control_steps": 2430, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1215, "magicsim/control_steps": 2430, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index ca2e2f9792f7225efa98b36ba7f353c3ab0a0ef4..c5f353e518de302b0bcc40366ec444d4d03f9352 100644 --- a/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2125/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2126/rec/check.json b/N053/go2_arm/dr_on/seed_2126/rec/check.json index 88a4f101b048cfc296b2340fce4e197cb6952352..2db1fc7be6dda16805a95b85e768b64e3f5a769f 100644 --- a/N053/go2_arm/dr_on/seed_2126/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2126/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2126/rec/check.txt b/N053/go2_arm/dr_on/seed_2126/rec/check.txt index d4c82a66317b37bea4f34f0bd4fe66682406ebaa..4e9db7f93595d4b2c166d2233ce614f8e1312eb2 100644 --- a/N053/go2_arm/dr_on/seed_2126/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2126/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2126_r1/out/rec/memorystage/0 --- episode file episode_000000.parquet: 1161 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1161 at dataset fps 25 = 46.44 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1161 rows vs ceil(2321/2) = 1161 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2381, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 20151b350b02d5bc84f089858d8d3c7de188480f..3b0859f609f650fa1ab4d9f932273b8ffce34841 100644 --- a/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1161, "magicsim/control_steps": 2321, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1161, "magicsim/control_steps": 2321, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index d335c713d64bd4258e198e4b27eb1de6cd736ce7..c9e4b627d0a7ab434c4ca2cc78e6e4a096109711 100644 --- a/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2126/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2127/rec/check.json b/N053/go2_arm/dr_on/seed_2127/rec/check.json index 7f8d6bf8a1da65c4e4843e4caf50dba37a4fbd49..449804f906a8b4a685749a21bdc68b111791cf42 100644 --- a/N053/go2_arm/dr_on/seed_2127/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2127/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2127/rec/check.txt b/N053/go2_arm/dr_on/seed_2127/rec/check.txt index ddd2730ad6f3acfd12c6e8908c41f9f15b1777fc..7eff98363829c798366797d0ef11f0a32863b8de 100644 --- a/N053/go2_arm/dr_on/seed_2127/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2127/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2127/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1162 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1162 at dataset fps 25 = 46.48 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1162 rows vs ceil(2324/2) = 1162 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2383, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 30e72b7c8a1ebd8c09dcdd272150e01c50b11934..4894f4e893c348aa70dfc1bf737036b47795a328 100644 --- a/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1162, "magicsim/control_steps": 2324, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1162, "magicsim/control_steps": 2324, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 6b8eaaeab53c3e0793c0bc74029dbbab846c5804..8f75217392e676210f1ea3123040979abaddcaf3 100644 --- a/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2127/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2128/rec/check.json b/N053/go2_arm/dr_on/seed_2128/rec/check.json index 9919cfc48f3f4e5f658cf02174c176ba1d03e371..2972b98fe1df46231dc1c4aac4cbafb1f6ee3e47 100644 --- a/N053/go2_arm/dr_on/seed_2128/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2128/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2128/rec/check.txt b/N053/go2_arm/dr_on/seed_2128/rec/check.txt index 77440ccae16e7bd57fb32a14b056657d1dfea803..9cb73e8de9785a0e7f025f6308515b0ed013f9f2 100644 --- a/N053/go2_arm/dr_on/seed_2128/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2128/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2128/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1140 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1140 at dataset fps 25 = 45.60 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1140 rows vs ceil(2280/2) = 1140 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2339, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 4f11c42fb6f19f1a7dcd56924f58f59889164ba2..c03411e338f618506017cd1ca3318513fd45c2f2 100644 --- a/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1140, "magicsim/control_steps": 2280, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1140, "magicsim/control_steps": 2280, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index fddd2994e4c01857d9c0351114f29aeb3e03ecca..8580e52ca12d3bae2b7354e55b50d16376d52f93 100644 --- a/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2128/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2129/rec/check.json b/N053/go2_arm/dr_on/seed_2129/rec/check.json index 6e4b7a82d0c578f24644832d41590ee57eeb1123..a05f02205649505cef84f50b0591ce6f1fa8977a 100644 --- a/N053/go2_arm/dr_on/seed_2129/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2129/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2129/rec/check.txt b/N053/go2_arm/dr_on/seed_2129/rec/check.txt index 8a9c3a9eecc136c658b66bfb4f4a1fcfebfe7985..055b45f08e88180b740d3bed1516c9db2fa2097a 100644 --- a/N053/go2_arm/dr_on/seed_2129/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2129/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2129/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1219 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1219 at dataset fps 25 = 48.76 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1219 rows vs ceil(2437/2) = 1219 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2497, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index a89e9e7021f7e2adc9d9d1898d2c7ff2fbc0b99a..6a31eb077d723b8ce418bf5a12bb9a3e81047501 100644 --- a/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1219, "magicsim/control_steps": 2437, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1219, "magicsim/control_steps": 2437, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index d858aaec9ac42610663d673e06528ad768227935..6c6f4354f9716bf69c7fd2a5a77118f09ea68ee8 100644 --- a/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2129/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2130/rec/check.json b/N053/go2_arm/dr_on/seed_2130/rec/check.json index b9398dd27e98b054fdb9150bcbd27d636d17d5ba..522e1ada5035a77dd721e8f6e01cd0b4d10c7703 100644 --- a/N053/go2_arm/dr_on/seed_2130/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2130/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2130/rec/check.txt b/N053/go2_arm/dr_on/seed_2130/rec/check.txt index 8d6971a168adcfe14ed429589847b9ca139fb8f2..9735c6724bb638e14ee5db6dfaaff49b50ca3a97 100644 --- a/N053/go2_arm/dr_on/seed_2130/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2130/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2130/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1330 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1330 at dataset fps 25 = 53.20 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1330 rows vs ceil(2660/2) = 1330 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2719, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 2272951461b0d2c02265343c9804e17925648f0a..429b91a1358083fdfecdaae5b535dde09d20c81c 100644 --- a/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1330, "magicsim/control_steps": 2660, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1330, "magicsim/control_steps": 2660, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is on top of the rack. The white st booth r is to the left of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 1e43bebab0870f393dd19db0861d9359b8b24ad8..76e0c7ed12c1f13851b0276ac9160edf7319f631 100644 --- a/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2130/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2131/rec/check.json b/N053/go2_arm/dr_on/seed_2131/rec/check.json index e3bc58b34e233dcc0f515de754665504c7d4701d..a26e02664406008ec1b24f59fa6941901f1f145a 100644 --- a/N053/go2_arm/dr_on/seed_2131/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2131/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2131/rec/check.txt b/N053/go2_arm/dr_on/seed_2131/rec/check.txt index 45bde6985653d6b042848e3f91706b4ac9188c54..990ca0c78ff293ae2f9e3964d8c5c1b4eebecd46 100644 --- a/N053/go2_arm/dr_on/seed_2131/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2131/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2131/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1306 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1306 at dataset fps 25 = 52.24 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1306 rows vs ceil(2611/2) = 1306 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2671, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index ef07d3e492f1d8a83f6e617a8dc5a234bfdbece2..2aac9192a9bd557ba1e734f85360d360ab5b92e7 100644 --- a/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1306, "magicsim/control_steps": 2611, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1306, "magicsim/control_steps": 2611, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 8c36deb907ac2ed156cd747086010651938571d4..8351c0924401d031ee89b857fed4691060e9f1e3 100644 --- a/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2131/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2132/rec/check.json b/N053/go2_arm/dr_on/seed_2132/rec/check.json index a621106431c37a4a2ad6baa6412c0a73325b6c8a..f5ad813ddd1e05a07ab9db307afd88854eac2fea 100644 --- a/N053/go2_arm/dr_on/seed_2132/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2132/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2132/rec/check.txt b/N053/go2_arm/dr_on/seed_2132/rec/check.txt index fd0e29ed289f9b61219fdab5de90a9998b46b28a..4532515637b35fa16f0cb58875e6f8c425e17dfe 100644 --- a/N053/go2_arm/dr_on/seed_2132/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2132/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2132/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1210 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1210 at dataset fps 25 = 48.40 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1210 rows vs ceil(2420/2) = 1210 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2479, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 2ed08c816b3fc13416c67ecd2f778a321f0acc5e..53b46432d6e2256565fd85c289b2bdbe3868ec7b 100644 --- a/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1210, "magicsim/control_steps": 2420, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1210, "magicsim/control_steps": 2420, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index b65d50e6fbd1259cc1d94a126a4364eee0da5fcf..0a748523458e535d138c54b2e0481a0a7d6c30a2 100644 --- a/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2132/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2133/rec/check.json b/N053/go2_arm/dr_on/seed_2133/rec/check.json index 6c62a3618dc3fbc0f5edadf9c78306cd92c88e16..c1093be2cb64aa4e58cd7ff2830599860429e0dd 100644 --- a/N053/go2_arm/dr_on/seed_2133/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2133/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2133/rec/check.txt b/N053/go2_arm/dr_on/seed_2133/rec/check.txt index 6c5d45abc69bb85086ee5f1d2f28edf29a9b1b9d..093d6c888524f755de61181c84713adc2a077f6e 100644 --- a/N053/go2_arm/dr_on/seed_2133/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2133/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2133/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1250 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1250 at dataset fps 25 = 50.00 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1250 rows vs ceil(2499/2) = 1250 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2559, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 2ea9e7b38f3aed632a210492e96ae52f0730a107..249914cd5eb302d45f741582e97769ca9326e82c 100644 --- a/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1250, "magicsim/control_steps": 2499, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1250, "magicsim/control_steps": 2499, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 4b854b8ea2125e4a5af38c6e1b9e196b06b154c8..57e9c22dad4a68eacfa82e1f20f6b33b7d38200c 100644 --- a/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2133/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2134/rec/check.json b/N053/go2_arm/dr_on/seed_2134/rec/check.json index 00771595606da4191e963798058409bd30af025f..1e6d2816a016f56681b7e5b81aab12aa6c5a38ab 100644 --- a/N053/go2_arm/dr_on/seed_2134/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2134/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2134/rec/check.txt b/N053/go2_arm/dr_on/seed_2134/rec/check.txt index ee0d27ff2303b08d1ccafa7986b6d29699d8e546..94ab2112f18be1e5284599db8720c5f3c86aa095 100644 --- a/N053/go2_arm/dr_on/seed_2134/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2134/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2134/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1619 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1619 at dataset fps 25 = 64.76 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1619 rows vs ceil(3238/2) = 1619 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3297, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 0a783776dcdabc40d9c39adc02cf1539bf0fb41c..1e4ca510254693f6122ee4f7b53a887fd12b686e 100644 --- a/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1619, "magicsim/control_steps": 3238, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1619, "magicsim/control_steps": 3238, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 24ce6f9739e79f3d2a0728de3814ba04c697f3ac..60d72f10dbbc574155a5e71875e883e0f38b40f2 100644 --- a/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2134/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2135/rec/check.json b/N053/go2_arm/dr_on/seed_2135/rec/check.json index 33bc759314a2188dfab558fa7e3793277dc9ba29..401207c7fa70db73fd91d85112d16807b592385c 100644 --- a/N053/go2_arm/dr_on/seed_2135/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2135/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2135/rec/check.txt b/N053/go2_arm/dr_on/seed_2135/rec/check.txt index 46e9b2f9950b1897e6e17a51dc5c7d03d3a0b1a2..6c6e8dee4821881663d7d698cda33c854b3c56e0 100644 --- a/N053/go2_arm/dr_on/seed_2135/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2135/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2135/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1152 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1152 at dataset fps 25 = 46.08 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1152 rows vs ceil(2304/2) = 1152 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2363, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index a0b310a51da77258b7e29774202316b296b5e285..6fa111acf265a92e059f25d169c2ce87081bb0e4 100644 --- a/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1152, "magicsim/control_steps": 2304, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1152, "magicsim/control_steps": 2304, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index e4fca0ed83591e8282d28fd467bb60f98caedfbc..3ca2bb1e10f06a9a017c9afb5e76e10cf5ddfd6e 100644 --- a/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2135/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2136/rec/check.json b/N053/go2_arm/dr_on/seed_2136/rec/check.json index f0d35cdd28ea4ed035002e0448f8f833ea0055bb..8a1468e95f4971d90aaef1fa5dc711d7136e5e45 100644 --- a/N053/go2_arm/dr_on/seed_2136/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2136/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2136/rec/check.txt b/N053/go2_arm/dr_on/seed_2136/rec/check.txt index b2acd512fb219047a66e92a5603b58bfb1e3ed29..1117466daa919cd2ad9f5620c0945cbe646b5946 100644 --- a/N053/go2_arm/dr_on/seed_2136/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2136/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2136/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1181 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1181 at dataset fps 25 = 47.24 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1181 rows vs ceil(2362/2) = 1181 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2421, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index c2d64c6240a9231c10a8b8bafcbae864208039cb..69e1654e00fa1faf4f475e896408feda3f2678f3 100644 --- a/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1181, "magicsim/control_steps": 2362, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1181, "magicsim/control_steps": 2362, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is behind the rack. The white st booth r is on top of the rack. The white st booth r is in front of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index bf9993d34f46f5f2425a8c82523aa485bf4cb112..010832ba0569c2645e6786ceb63c0ad6ce9b2bf2 100644 --- a/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2136/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2137/rec/check.json b/N053/go2_arm/dr_on/seed_2137/rec/check.json index 5f9aff89f258c794c924cbe759a99191db6d8a2f..e6790dba4f496e206d995090a1178bfdf4f1402d 100644 --- a/N053/go2_arm/dr_on/seed_2137/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2137/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2137/rec/check.txt b/N053/go2_arm/dr_on/seed_2137/rec/check.txt index 14554824ae17b0faf15434165f1da85fc053c0af..23750eec04d9ea8e2caa1e72658eeea932691bf0 100644 --- a/N053/go2_arm/dr_on/seed_2137/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2137/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2137/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1214 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1214 at dataset fps 25 = 48.56 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1214 rows vs ceil(2428/2) = 1214 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2487, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 73b5a11c1cfc3537b7a9216ba34aa1ac6dc8def1..39be787e2eff7a9a2ab228afcf0b596decdc1073 100644 --- a/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1214, "magicsim/control_steps": 2428, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1214, "magicsim/control_steps": 2428, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 916b09b9e54b50a9425890c633f372036f320aec..63e894c8321764d7fd8eea98705fb6a8ae9a29c2 100644 --- a/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2137/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2138/rec/check.json b/N053/go2_arm/dr_on/seed_2138/rec/check.json index 3ff4e51a73d68a1577a01be8fab65879f63c2aee..e357429e3c66868b422997dfa5783c1e176e235a 100644 --- a/N053/go2_arm/dr_on/seed_2138/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2138/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2138/rec/check.txt b/N053/go2_arm/dr_on/seed_2138/rec/check.txt index a90677efd777c1ebcd078ef26ad87ed238ab5bf5..ce55c8186372e1955d290532764d7410d799fa0d 100644 --- a/N053/go2_arm/dr_on/seed_2138/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2138/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2138/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1200 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1200 at dataset fps 25 = 48.00 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1200 rows vs ceil(2399/2) = 1200 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2459, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 6a257f5c1d8b606dc59dd5684d8ef0bb2d87a927..35c9ef8e9d118a2490855143b6a4eca555da8a0b 100644 --- a/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1200, "magicsim/control_steps": 2399, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1200, "magicsim/control_steps": 2399, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 692f0f193cc7d23af7e00a187b17acbe23a1eaca..f4ad205be4fe935fb0498a0d1e9dd7fb03babbf9 100644 --- a/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2138/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/go2_arm/dr_on/seed_2139/rec/check.json b/N053/go2_arm/dr_on/seed_2139/rec/check.json index aa6c224a7e7c40471667d2ad5f980d1fcde21a15..f741d922c126971448d0e689a4b53ec8608c46d6 100644 --- a/N053/go2_arm/dr_on/seed_2139/rec/check.json +++ b/N053/go2_arm/dr_on/seed_2139/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/go2_arm/dr_on/seed_2139/rec/check.txt b/N053/go2_arm/dr_on/seed_2139/rec/check.txt index a7c1980b19cf8b766d9fc9304fe9f8851690ad2d..4ae3d18f0dced947329ff777824d4723df133915 100644 --- a/N053/go2_arm/dr_on/seed_2139/rec/check.txt +++ b/N053/go2_arm/dr_on/seed_2139/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_go2_on_2139/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1228 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 26, info shape [26], 26 names first: ['joint_0', 'joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5'] [PASS] action shape: parquet width 12, info shape [12], 12 names first: ['base_action_0', 'base_action_1', 'base_action_2', 'base_action_3', 'eef0.pos_x', 'eef0.pos_y'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1228 at dataset fps 25 = 49.12 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1228 rows vs ceil(2456/2) = 1228 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2515, step differences [2] diff --git a/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 3daba863e5389c4316f2dd4f6c85aa02101bdedb..ef13369514afe0a6f95b0762a3479b2dcf0c4b55 100644 --- a/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1228, "magicsim/control_steps": 2456, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1228, "magicsim/control_steps": 2456, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "An Unitree Go2 EDU quadruped with a Piper arm is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is on top of the rack. The white st booth r is to the right of the rack. The rack is on top of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is in front of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from front-left, at table height."} diff --git a/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index c389ce40cf3de2099406cf6b13a7514390a01d45..86a6cbcfbcb7ba88a12ad7d1d4956db7806c4aed 100644 --- a/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/go2_arm/dr_on/seed_2139/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1001/rec/check.json b/N053/sawyer_umi/dr_off/seed_1001/rec/check.json index f214e931f15b68cb11e81bfe8fe7f240e73e2ce0..c45bfb01da073cfdfd56b73f9b95b9f98a154fb2 100644 --- a/N053/sawyer_umi/dr_off/seed_1001/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1001/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1001/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1001/rec/check.txt index 0f39bb62532e2c98ad6e190b68641de7113c0101..35c2cd25fe2bf23a6dd590d36ffe652145e5f816 100644 --- a/N053/sawyer_umi/dr_off/seed_1001/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1001/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1001/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1598 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1598 at dataset fps 25 = 63.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1598 rows vs ceil(3196/2) = 1598 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3255, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 5a2ff98d582928193fd3172f135f5b1ac9c0bf6c..66be70dbd1cee0d8d795aeaa4b9ca7d8655dcd21 100644 --- a/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1598, "magicsim/control_steps": 3196, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1598, "magicsim/control_steps": 3196, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 9366019a34014b6136261e1617992692e1eddb47..ef38f3aefcfdc62027aab3013f92b0be6b3289e4 100644 --- a/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1002/rec/check.json b/N053/sawyer_umi/dr_off/seed_1002/rec/check.json index 2ead09db3737dba389bcbcbdbfd0b3f8043661b7..64022ee644e7e3847682a738c061d1104a08f9de 100644 --- a/N053/sawyer_umi/dr_off/seed_1002/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1002/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1002/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1002/rec/check.txt index 9fa18965dbf17dc4394bc9e32af13dc491696762..486fbaa1f4c1c82cd8bfdb0fe9e65b9efab2a221 100644 --- a/N053/sawyer_umi/dr_off/seed_1002/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1002/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1002/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1671 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1671 at dataset fps 25 = 66.84 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1671 rows vs ceil(3342/2) = 1671 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3401, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 47c83271dfc89efbe8c008b2863af7ca5e23e512..f8018e66724c83dc12057eb5812075212eba09a8 100644 --- a/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1671, "magicsim/control_steps": 3342, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1671, "magicsim/control_steps": 3342, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index f57294895ff9c5b47b6b953ce22f50412cc4c68e..9abaf1265e6970f723cbd9f7f053abf44222d166 100644 --- a/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1003/rec/check.json b/N053/sawyer_umi/dr_off/seed_1003/rec/check.json index 23958c74de493af0b1de6b2eabc938e5d3db133a..4470817f8e612079b1c6a57f989c615159812d29 100644 --- a/N053/sawyer_umi/dr_off/seed_1003/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1003/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1003/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1003/rec/check.txt index 81bccee714ccfe0258c511ccc2718d425bdb13b3..31cc8cb652a34b5a011414c61945f472cb5d6986 100644 --- a/N053/sawyer_umi/dr_off/seed_1003/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1003/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1003/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1476 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1476 at dataset fps 25 = 59.04 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1476 rows vs ceil(2951/2) = 1476 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3011, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index ed385a09bcd24b2c68a4e8b7b085f3131b4ecef3..cc1bee66f637917427c5850aa3ce478f94204d1f 100644 --- a/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1476, "magicsim/control_steps": 2951, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1476, "magicsim/control_steps": 2951, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index cc93ead11c866ddbc18b8f2111ea007074caaed1..4a0a1abb6862d0a8b5a9478bcdc27611b5dd64bb 100644 --- a/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1004/rec/check.json b/N053/sawyer_umi/dr_off/seed_1004/rec/check.json index cd8bea9ae0a54300f4c028e59df5319ae947b21c..99bbd01b2cb9ed06ad80cb2c4bd7b5274eb853b4 100644 --- a/N053/sawyer_umi/dr_off/seed_1004/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1004/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1004/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1004/rec/check.txt index 99818b707ca437c7c808553f9f67622a16626691..7107f5a51af8fe54d726de145b9f9d5300d367be 100644 --- a/N053/sawyer_umi/dr_off/seed_1004/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1004/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1004/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1613 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1613 at dataset fps 25 = 64.52 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1613 rows vs ceil(3225/2) = 1613 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3285, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 61030a520fdc3c5e20f06fe4e20ea16d14841b49..f7e32e01279aef41e890cc534b2ec29f7299e406 100644 --- a/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1613, "magicsim/control_steps": 3225, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1613, "magicsim/control_steps": 3225, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 8e08d72752f321cc7fdb5a437ab653e433f0bb32..3442ad312e2cb147ae5b08aaaf389692e2a70a37 100644 --- a/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1005/rec/check.json b/N053/sawyer_umi/dr_off/seed_1005/rec/check.json index 16495f4e960cd89dfdb3be90416f28eb7ad06293..ba0a03671abee32ce332c409fc20fe80a4c8e1c2 100644 --- a/N053/sawyer_umi/dr_off/seed_1005/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1005/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1005/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1005/rec/check.txt index 0d7fc461bec94d69cca676fd407eba73c5f2fe07..78f24084442ccba0df94ec7b2edc1373e63f36bb 100644 --- a/N053/sawyer_umi/dr_off/seed_1005/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1005/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1005/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1469 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1469 at dataset fps 25 = 58.76 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1469 rows vs ceil(2937/2) = 1469 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2997, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 555c5c5b5ded23a5909743c63b7e3bddf06e3ca3..faf80069dd8447ff857012392cb60bf8098a498a 100644 --- a/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1469, "magicsim/control_steps": 2937, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1469, "magicsim/control_steps": 2937, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 77bd51eeeae3230f6c864918d3cc0959079cdaa3..6234a65354b8703891795305bcb0bcfac6fa9759 100644 --- a/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1006/rec/check.json b/N053/sawyer_umi/dr_off/seed_1006/rec/check.json index 31b1064103fc2806508d8b569318c74061acb9d2..6d5a1e1aff83dfd43bf05c424e0aed5420f9e878 100644 --- a/N053/sawyer_umi/dr_off/seed_1006/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1006/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1006/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1006/rec/check.txt index 90841ad4ad97d5e37021573af183999b1b0fe2bd..ff46cbe388ccff65d06ce5b48510ac8253b8ae28 100644 --- a/N053/sawyer_umi/dr_off/seed_1006/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1006/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1006/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1647 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1647 at dataset fps 25 = 65.88 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1647 rows vs ceil(3294/2) = 1647 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3353, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 9424c1115eb34df328b96856026a142d4b69ab88..131e3b513b109e1345997394b1eab3d7c83b3858 100644 --- a/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1647, "magicsim/control_steps": 3294, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1647, "magicsim/control_steps": 3294, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 54393d53d67e31a44385653940be9c9c867d601e..f9721e8bb9793e9fa418412058fe8d7718f964b1 100644 --- a/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1007/rec/check.json b/N053/sawyer_umi/dr_off/seed_1007/rec/check.json index dd4bf733633c24e3a3ef91cd895b79c9dbe87b3f..2605f813bb257c909c823673b951ff3d675ccedb 100644 --- a/N053/sawyer_umi/dr_off/seed_1007/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1007/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1007/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1007/rec/check.txt index 6ccb76618bfd5df6b9f9486ae922ff24fe0f83c9..2a2d6c4561f8afea596f701d1e7fe7ad2afcc811 100644 --- a/N053/sawyer_umi/dr_off/seed_1007/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1007/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1007/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1475 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1475 at dataset fps 25 = 59.00 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1475 rows vs ceil(2949/2) = 1475 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3009, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 937f70b7c72702d0c1c238a643015116328a7e24..8dccde0f7516786a6bf96fc86e4143ea5bc399a5 100644 --- a/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1475, "magicsim/control_steps": 2949, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1475, "magicsim/control_steps": 2949, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index ade82ccd03fa8defa2491135bac6ccd1ad8ec13c..becc76314d004ba6b1453c1faa500fef84586884 100644 --- a/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1008/rec/check.json b/N053/sawyer_umi/dr_off/seed_1008/rec/check.json index 918fa30113d536ccc6038a43abc2af13ec9513da..8de67ccc3a339584bf017e028dea2b95a62c8de6 100644 --- a/N053/sawyer_umi/dr_off/seed_1008/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1008/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1008/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1008/rec/check.txt index c6bea7dadadfdc65c155e0b8d42350e1a66ead5f..77f3248fd51d97f190b7535d65393ab89ce9c43c 100644 --- a/N053/sawyer_umi/dr_off/seed_1008/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1008/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1008/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1457 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1457 at dataset fps 25 = 58.28 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1457 rows vs ceil(2913/2) = 1457 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2973, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 435366c0d8a422ee725d51a761e7054c9154b4de..99cc8607d576721964481f0059923cd164c33fc7 100644 --- a/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1457, "magicsim/control_steps": 2913, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1457, "magicsim/control_steps": 2913, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 34bf8ccdea561f18ee01ff95784a5cd8bd9d3ae5..beed257919b59d6fe0b65ccfb2b041392b6dc71f 100644 --- a/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1009/rec/check.json b/N053/sawyer_umi/dr_off/seed_1009/rec/check.json index 52d770d0279af3a985a592c73a7d584be9521341..d877405067a140107b86fc907b61a1a03c757156 100644 --- a/N053/sawyer_umi/dr_off/seed_1009/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1009/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1009/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1009/rec/check.txt index 2245b9b4a3cd24372b5c5956f7d69a3dcd5c05d5..817d488c8a8412681d33cbabc970173f936c5483 100644 --- a/N053/sawyer_umi/dr_off/seed_1009/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1009/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1009/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1645 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1645 at dataset fps 25 = 65.80 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1645 rows vs ceil(3290/2) = 1645 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3349, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index c40c5be2e1f2f129831124b9e619a443396d6684..4cc1d692ffd71088ebb5e2ed1d7c465c262213d9 100644 --- a/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1645, "magicsim/control_steps": 3290, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1645, "magicsim/control_steps": 3290, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index a26c53265ffab4b3a1879a751d788e9c3fe33c28..6203d4a741e9765fed78be691b27910ff7cc6ddc 100644 --- a/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1009/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1010/rec/check.json b/N053/sawyer_umi/dr_off/seed_1010/rec/check.json index 7c4640b5e52140445aa48084020515279088c5ac..59c4f057bb77f03971b0e9fc08d28c36abb82dc4 100644 --- a/N053/sawyer_umi/dr_off/seed_1010/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1010/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1010/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1010/rec/check.txt index 2b50f9da97993e252cc0b77338f4766de173a023..6543adac2b12d740048346e7cfec69cf4f14e846 100644 --- a/N053/sawyer_umi/dr_off/seed_1010/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1010/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1010/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1673 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1673 at dataset fps 25 = 66.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1673 rows vs ceil(3345/2) = 1673 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3405, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index b58f61e6e7589a35a5ec48591f58275ae0ab7e95..fcc7fd0bc4fd755b4c81ad87ebc8a859a2866af9 100644 --- a/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1673, "magicsim/control_steps": 3345, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1673, "magicsim/control_steps": 3345, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index bc54b1a928d13900d1772411964b336f01f35be7..14d310d466d344fb2f22ced807f334695775d921 100644 --- a/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1010/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1011/rec/check.json b/N053/sawyer_umi/dr_off/seed_1011/rec/check.json index b83f07084221e265516c7a680c639a147b2d527a..d61e8a334fb13547409a76c672c5fda24a930814 100644 --- a/N053/sawyer_umi/dr_off/seed_1011/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1011/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1011/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1011/rec/check.txt index 88ac5392339808f5517a9aec0303fa5c4e1c1184..f29a42eb89e8d24502de6e6b2f4a7d4bc6751f43 100644 --- a/N053/sawyer_umi/dr_off/seed_1011/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1011/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1011/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1446 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1446 at dataset fps 25 = 57.84 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1446 rows vs ceil(2891/2) = 1446 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2951, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 6d7b92606ebd39733a4584594df555404773b432..1a8b298bb33c668f59eba8962fe14a007c2cfcc7 100644 --- a/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1446, "magicsim/control_steps": 2891, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1446, "magicsim/control_steps": 2891, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 595caa3e8b89d2751c3d2ebf522e74ee4b174f8e..26eaf9351a50715c7959666e8bd85fb6472e2d54 100644 --- a/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1011/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1012/rec/check.json b/N053/sawyer_umi/dr_off/seed_1012/rec/check.json index 76c23d54a234e805c5a96e07f5d22e0d90ed91c3..9970940685259fcf653217a872d671658aa3db3f 100644 --- a/N053/sawyer_umi/dr_off/seed_1012/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1012/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1012/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1012/rec/check.txt index d0e08fd01ab5d824d32b83594d1b69c104863edc..e7f2109553c5cb0cdcb17072c9fdc95f9f96b8e3 100644 --- a/N053/sawyer_umi/dr_off/seed_1012/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1012/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1012/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1657 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1657 at dataset fps 25 = 66.28 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1657 rows vs ceil(3314/2) = 1657 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3373, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 0a889b46de89eb2189897181a03c553acbd132a6..e7271664822d402334fb33641ec932f5e60227c3 100644 --- a/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1657, "magicsim/control_steps": 3314, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1657, "magicsim/control_steps": 3314, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 8f652e3fcea49d816e33a3ceb298ed115655932c..435754cda5785fdd1c4de57284ec4eb823322f83 100644 --- a/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1012/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1013/rec/check.json b/N053/sawyer_umi/dr_off/seed_1013/rec/check.json index 9d99ccca2158235e44a4f203d589bcec6640bc08..dc904d1c478ca89a0dbc5a8142dab4102853ec0a 100644 --- a/N053/sawyer_umi/dr_off/seed_1013/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1013/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1013/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1013/rec/check.txt index 1d67bbb73211f811b5e204cf5016a4577f2273bb..25cc706e38eb929022b836713dbfb593287a56aa 100644 --- a/N053/sawyer_umi/dr_off/seed_1013/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1013/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1013_r1/out/rec/memorystage/ --- episode file episode_000000.parquet: 1522 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1522 at dataset fps 25 = 60.88 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1522 rows vs ceil(3043/2) = 1522 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3103, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 34305b7570c043e1e0cf8a9f1488e64f2bac0294..6e6d21277a48fd74cc6b7b8bb992a4702616a0e2 100644 --- a/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1522, "magicsim/control_steps": 3043, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1522, "magicsim/control_steps": 3043, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 9d0e2cd766eb7362ebd48dd424e33771d061e227..863db3c7a79dce7187c20766f91a404b7bc5e642 100644 --- a/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1013/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1014/rec/check.json b/N053/sawyer_umi/dr_off/seed_1014/rec/check.json index 0c365eaa178ea66de6e375c31a006c4985fb6a73..a6cecbde74f3422026ebbd4a3f4f586b03a686bf 100644 --- a/N053/sawyer_umi/dr_off/seed_1014/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1014/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1014/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1014/rec/check.txt index 92b57d5ed57af902a8b15abb21aa808e962ef62c..8866a72ddee0dc3820a6695848d5383d4fb532a9 100644 --- a/N053/sawyer_umi/dr_off/seed_1014/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1014/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1014/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1687 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1687 at dataset fps 25 = 67.48 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1687 rows vs ceil(3374/2) = 1687 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3433, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 35cdb052b7b9f546632df2c1a181f77bce272d7b..f4cbff085782776fc54e5002856aaad421242290 100644 --- a/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1687, "magicsim/control_steps": 3374, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1687, "magicsim/control_steps": 3374, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 3e814261fffac5b1521edc2901b788ca0283aa54..8ec7080d779155cfc41c5e1c03a73a792789a5c9 100644 --- a/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1014/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1015/rec/check.json b/N053/sawyer_umi/dr_off/seed_1015/rec/check.json index 5d5b58a0a3d731402f4f11a3704a18b616dcb191..ae9261db58dd014dab46c8b9a883be1c03e6dd38 100644 --- a/N053/sawyer_umi/dr_off/seed_1015/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1015/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1015/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1015/rec/check.txt index 06cff869a9ffb55f290c1d1db22f1f7bba834b22..17a535f735ae8dc5ccd6dd33337b07f11edd8865 100644 --- a/N053/sawyer_umi/dr_off/seed_1015/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1015/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1015/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1654 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1654 at dataset fps 25 = 66.16 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1654 rows vs ceil(3307/2) = 1654 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3367, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index dfc16e67403954a7eb8900d8ba6b111954018e9f..8577752cab58f1b6c14f82e283fa80abae8dd660 100644 --- a/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1654, "magicsim/control_steps": 3307, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1654, "magicsim/control_steps": 3307, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 8a3bf032d7957cafbb399c13f4dc0fe14198328e..b50f9a63daeeca383e520335a6a4e3a990fa0607 100644 --- a/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1015/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1016/rec/check.json b/N053/sawyer_umi/dr_off/seed_1016/rec/check.json index c4a191509aa84abe2d2ee63792958429b83c8b0d..da7934eb9b077445563847ccd61e4c3630d965ac 100644 --- a/N053/sawyer_umi/dr_off/seed_1016/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1016/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1016/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1016/rec/check.txt index 392c36c7419c2193f15453a28cb13e5d0822e58a..e3a47f364d37a16945e5ca672e837e5e8403eb63 100644 --- a/N053/sawyer_umi/dr_off/seed_1016/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1016/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1016_r2/out/rec/memorystage/ --- episode file episode_000000.parquet: 1639 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1639 at dataset fps 25 = 65.56 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1639 rows vs ceil(3277/2) = 1639 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3337, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index bc46b605af60ea4396fc5693d7cc8669cef6df00..768b6b81afe3341c3695140cd333e36de0043a30 100644 --- a/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1639, "magicsim/control_steps": 3277, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1639, "magicsim/control_steps": 3277, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 862f9b8475bb1e64bd667db68a0eab1627684124..d719ce8c605547068f6489fbac1bbc765545e8e9 100644 --- a/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1016/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1017/rec/check.json b/N053/sawyer_umi/dr_off/seed_1017/rec/check.json index a90fd118fd1786b31728a72d72a6aff51d3439ef..4a1d867b79bb1105e65ba7fd0a1f39d0da673a1b 100644 --- a/N053/sawyer_umi/dr_off/seed_1017/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1017/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1017/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1017/rec/check.txt index 4580e4fa53497c678a26ddf6029adb80da1780f2..55d8d97b54ba75f328b5ffcbe6566f10eefe9817 100644 --- a/N053/sawyer_umi/dr_off/seed_1017/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1017/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1017_r2/out/rec/memorystage/ --- episode file episode_000000.parquet: 1693 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1693 at dataset fps 25 = 67.72 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1693 rows vs ceil(3385/2) = 1693 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3445, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 1741be027e2308b068337d40059951403fcb59d4..bd351a2cd61959286d2fcb7a56ddb03bbde0c857 100644 --- a/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1693, "magicsim/control_steps": 3385, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1693, "magicsim/control_steps": 3385, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index af516511d0b76321ccbef49e87e639e1a92ba9c5..dfa030074a787d447da5cfa902e7e5abf4cd994f 100644 --- a/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1017/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1018/rec/check.json b/N053/sawyer_umi/dr_off/seed_1018/rec/check.json index cc2f8a691333c625568d46e47d196201b38daaa9..2192156a5544d4430c608df5cc4d707c199f5234 100644 --- a/N053/sawyer_umi/dr_off/seed_1018/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1018/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1018/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1018/rec/check.txt index f623e94f53a6d7a6790b4576b35e04974eb5e205..e89039773a8ee2aaec04c290f5353c0a1d17f674 100644 --- a/N053/sawyer_umi/dr_off/seed_1018/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1018/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1018/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1620 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1620 at dataset fps 25 = 64.80 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1620 rows vs ceil(3239/2) = 1620 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3299, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 1f20d099f5cb6a5fd947d5e47ad157dcd7d3701a..a55335c4c25d18d5f134c54875c3f55fd1f02764 100644 --- a/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1620, "magicsim/control_steps": 3239, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1620, "magicsim/control_steps": 3239, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 4fd479e44f2aad9e6926584855f33c45e7218185..c8ba63359b2c88966d643e5575c54d12422552db 100644 --- a/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1018/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1019/rec/check.json b/N053/sawyer_umi/dr_off/seed_1019/rec/check.json index 73b9a42fe8b9fa3c100035f5d80b70653b6cc67b..f193040ea72a2b8ee7a3debf5ad60d35c363684c 100644 --- a/N053/sawyer_umi/dr_off/seed_1019/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1019/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1019/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1019/rec/check.txt index cdca3402996122be0341e8af98001fe7b35d7f5e..2f4cf741e0dba6d258d2569d1a85a44cffa45c7c 100644 --- a/N053/sawyer_umi/dr_off/seed_1019/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1019/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1019/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1459 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1459 at dataset fps 25 = 58.36 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1459 rows vs ceil(2918/2) = 1459 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2977, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 15a68b3826e02b2bc219bee3006d9787fda7c4cf..bfa521b7e4747a818b489555e48c61f119792f58 100644 --- a/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1459, "magicsim/control_steps": 2918, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1459, "magicsim/control_steps": 2918, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 56a8252874dfb2d93b49608f85c8276f9f184c35..f07e1f7f849754e74e798b9f0b44d6778cb96fcb 100644 --- a/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1019/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1020/rec/check.json b/N053/sawyer_umi/dr_off/seed_1020/rec/check.json index 879ea29f43d0c5cfbe8133fbcc63df00d41cd7c8..d0894af0ee4edc08f023b04772105ca8f1a37593 100644 --- a/N053/sawyer_umi/dr_off/seed_1020/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1020/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1020/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1020/rec/check.txt index bc66aa41a9bb832552c2e5e55719f8817de5bb9f..5aed28265d497e0c4cce91f1efe92441666f37c3 100644 --- a/N053/sawyer_umi/dr_off/seed_1020/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1020/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1020/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1654 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1654 at dataset fps 25 = 66.16 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1654 rows vs ceil(3308/2) = 1654 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3367, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index f08ba81ce1df8c62f13df77abcb6ee4b28af3c43..d74401fef9ba1b848d73e5245917547d7002263b 100644 --- a/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1654, "magicsim/control_steps": 3308, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1654, "magicsim/control_steps": 3308, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 82ddae0e510e74bcca8452b21c84019ace9c19ee..febadc3980d652fcafdf97db808bcbd77f0054bb 100644 --- a/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1020/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1021/rec/check.json b/N053/sawyer_umi/dr_off/seed_1021/rec/check.json index f1f28a74f1b83427341cb58a304db448da2f312a..3a6c55f50fb63a34cf36e3ad5f8c609ef240c5ba 100644 --- a/N053/sawyer_umi/dr_off/seed_1021/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1021/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1021/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1021/rec/check.txt index a78913e8f81853302dde00c29bc34974a95e3594..b65de5456123b390139ef9d537a242c385296adf 100644 --- a/N053/sawyer_umi/dr_off/seed_1021/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1021/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1021_r1/out/rec/memorystage/ --- episode file episode_000000.parquet: 1521 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1521 at dataset fps 25 = 60.84 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1521 rows vs ceil(3041/2) = 1521 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3101, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index f0ed85920bd914c2be0bd0f64f77b47c31c3aeb3..0669f7975deb81d87c077a9bff37a3f75e65f2ea 100644 --- a/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1521, "magicsim/control_steps": 3041, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1521, "magicsim/control_steps": 3041, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index dc2ec1236ef049917fa02b44a0154a0ec91ad267..37e4f80d269e2e8592fc9cdeacf37e118c39403e 100644 --- a/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1021/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1022/rec/check.json b/N053/sawyer_umi/dr_off/seed_1022/rec/check.json index c3bfd368b852cf51fffedd7fb9a3c719a5f8e062..772432927fc805f878fc5feb0cb9c82116d71e90 100644 --- a/N053/sawyer_umi/dr_off/seed_1022/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1022/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1022/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1022/rec/check.txt index 5db72c518cd5b31a760ab9417d5ce39ad75000af..d72b996326aee910aebeb5653b632ff2ea61466c 100644 --- a/N053/sawyer_umi/dr_off/seed_1022/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1022/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1022/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1627 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1627 at dataset fps 25 = 65.08 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1627 rows vs ceil(3253/2) = 1627 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3313, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index e887aeee3762cf89ba891214b9f5a9e225fb44e4..ba3be3847fe9bd0ba1e707e23f10a97aac0a20dc 100644 --- a/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1627, "magicsim/control_steps": 3253, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1627, "magicsim/control_steps": 3253, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index ef13edf74b34f7a5b3844b871e3208d268da10ea..6b787a23aa1a8f81773a9d77d858189c4e4085a1 100644 --- a/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1022/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1023/rec/check.json b/N053/sawyer_umi/dr_off/seed_1023/rec/check.json index f59f9b9fee705207946b6bf5beaaa86e21848cad..db827d054f4a189c28fcbfbf7a773dcde38aa2e1 100644 --- a/N053/sawyer_umi/dr_off/seed_1023/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1023/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1023/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1023/rec/check.txt index bbab4a911acda3d13e43b573ff069af82e001a41..c6b9de58f39bab3bb7b9763b05912846c9dda0d3 100644 --- a/N053/sawyer_umi/dr_off/seed_1023/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1023/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1023/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1527 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1527 at dataset fps 25 = 61.08 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1527 rows vs ceil(3054/2) = 1527 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3113, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 6bc98e51d830ddc53b3fa9a42e008ee0b1005d23..1a8cde5bc0340df8f8af8fb9b2671587166d3407 100644 --- a/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1527, "magicsim/control_steps": 3054, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1527, "magicsim/control_steps": 3054, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index e6527c4f186cddccabbacbc64f2cb8faff9bb96e..c9b00dca1faed66adc016e31004421c6ebd6360c 100644 --- a/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1023/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1024/rec/check.json b/N053/sawyer_umi/dr_off/seed_1024/rec/check.json index 95d62f4573c6e61ea961c444eb9365ab51c265fd..759539775aae76ac88b0c4f79280d755883eb7ef 100644 --- a/N053/sawyer_umi/dr_off/seed_1024/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1024/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1024/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1024/rec/check.txt index 79eb53c457a0ab97df7111477cf0df8325d8c9e5..7b3b2bcbcee967d48e4783f2d4fd55a80e1a0e28 100644 --- a/N053/sawyer_umi/dr_off/seed_1024/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1024/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1024/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1649 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1649 at dataset fps 25 = 65.96 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1649 rows vs ceil(3297/2) = 1649 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3357, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 3c00e1533ead1ad082813973b49cbb44332c2053..f82707e5dc5fada1e2076434610629a054651713 100644 --- a/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1649, "magicsim/control_steps": 3297, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1649, "magicsim/control_steps": 3297, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index cd4f3445c7584bf792268e19d8281fdfb9bd3851..97bc663362ea238f3982561d424436406e9521cf 100644 --- a/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1024/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1025/rec/check.json b/N053/sawyer_umi/dr_off/seed_1025/rec/check.json index 1b2175fa5e86f89128ffcc518f45fa027ca7a50a..281cb9a9868ed8b1a1fee421f5566cc93a0f7bd9 100644 --- a/N053/sawyer_umi/dr_off/seed_1025/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1025/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1025/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1025/rec/check.txt index b87286e7d7deae1e55dba0641f6504f0db82fb64..ba0f2599d563313908815bdc0fdac0266ab33984 100644 --- a/N053/sawyer_umi/dr_off/seed_1025/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1025/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1025/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1683 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1683 at dataset fps 25 = 67.32 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1683 rows vs ceil(3365/2) = 1683 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3425, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index c7c3543181eedf9d018e8457c98d496948f6fc94..54284f24b7291872920b45a7afb34844aea1bdf0 100644 --- a/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1683, "magicsim/control_steps": 3365, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1683, "magicsim/control_steps": 3365, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index dab7eafe64aac3344bc6fd4c16eb531ddfbf7cd4..2bcde39ed5fd22f2dc1161da6e4aafa59b03cda2 100644 --- a/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1025/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1026/rec/check.json b/N053/sawyer_umi/dr_off/seed_1026/rec/check.json index 188ed769ed3ae73b871da1f70f1328561ede1ee8..2152682ff5859c61851d730f7de31fb3919c0bc0 100644 --- a/N053/sawyer_umi/dr_off/seed_1026/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1026/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1026/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1026/rec/check.txt index 9bbd9c531b512109870e205f5ba8903e68611d9a..6a8988cef1f0254c5c1972ee049ba166959c1d4d 100644 --- a/N053/sawyer_umi/dr_off/seed_1026/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1026/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1026/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1664 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1664 at dataset fps 25 = 66.56 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1664 rows vs ceil(3327/2) = 1664 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3387, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 4ca3b21e93ffd9b9b1196a951555ab55bfb26bae..dc9e7874e59fd3154f1099815ba66d9d69291881 100644 --- a/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1664, "magicsim/control_steps": 3327, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1664, "magicsim/control_steps": 3327, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 9a7b958aa2b385e2ffeb6f480169bc068986f6ad..05ae41d0f2f15ce77b176d1595c8ac0dcad1584c 100644 --- a/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1026/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1027/rec/check.json b/N053/sawyer_umi/dr_off/seed_1027/rec/check.json index 86a59542fb9d5c0fb13cc4063938f761122c794c..2bdc40512ac97cba54a03b144879bcd8b825c2c5 100644 --- a/N053/sawyer_umi/dr_off/seed_1027/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1027/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1027/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1027/rec/check.txt index 20104e7c2d862b4104f8a1755bbc471d5db7558c..64e6aa4d3ffe68d1339fb062cd22a8f68224e676 100644 --- a/N053/sawyer_umi/dr_off/seed_1027/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1027/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1027/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1627 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1627 at dataset fps 25 = 65.08 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1627 rows vs ceil(3253/2) = 1627 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3313, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index f98db52656c21aa9e8a2af7fdde35aeace1a286e..d416a054e06108d593ceedb7f37d53b69ce74737 100644 --- a/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1627, "magicsim/control_steps": 3253, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1627, "magicsim/control_steps": 3253, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 4e5c67ad37c2c27bbe2fb4b962015789b7c298a8..0e58f0b183f398c54d679898a46a391a352c9c73 100644 --- a/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1027/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1028/rec/check.json b/N053/sawyer_umi/dr_off/seed_1028/rec/check.json index 58314e7d54ccec3c6f6743a96c4d8666ddc5781b..c37cca43c9381b8cf963c216b3e96f136d350f51 100644 --- a/N053/sawyer_umi/dr_off/seed_1028/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1028/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1028/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1028/rec/check.txt index 5abea096b6ef0534f007e8ebda2a0412743ae33e..7966f741af12fc9f3d028480dd32fbe3f0f50c66 100644 --- a/N053/sawyer_umi/dr_off/seed_1028/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1028/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1028/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1653 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1653 at dataset fps 25 = 66.12 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1653 rows vs ceil(3306/2) = 1653 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3365, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 3ed3941558cee4794689cd54a22ab1f14760a86a..d009cf985b20d6c452b4afbabf0496b9d46218eb 100644 --- a/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1653, "magicsim/control_steps": 3306, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1653, "magicsim/control_steps": 3306, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 0b99405b2b4cbd29b011b9372cb418cfdbeeea39..a5912554261965b75f5bb8f1c79c1b894c765725 100644 --- a/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1028/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1029/rec/check.json b/N053/sawyer_umi/dr_off/seed_1029/rec/check.json index 06100a7727f612d3ad1be617b897273ab3c6a25a..e14eb60be761da3cb66fe40eeece6cdd916b98a3 100644 --- a/N053/sawyer_umi/dr_off/seed_1029/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1029/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1029/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1029/rec/check.txt index f359c06e99c43b4ca77374d6452a4c9a397e43a5..e6fd2fc19834d4f51e82b0d4af9d81dc279445b0 100644 --- a/N053/sawyer_umi/dr_off/seed_1029/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1029/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1029/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1666 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1666 at dataset fps 25 = 66.64 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1666 rows vs ceil(3331/2) = 1666 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3391, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 6d26a2cb63504006d1e732d4ca0d50cb8ed3f8b7..150bbd33e3e0e733d508d6833521523516c6db78 100644 --- a/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1666, "magicsim/control_steps": 3331, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1666, "magicsim/control_steps": 3331, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index dde8f8266ce22908ecc4c568aba041908f11fed6..f20b8c1bbb2281e300d5fb62415991bbec670151 100644 --- a/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1029/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1030/rec/check.json b/N053/sawyer_umi/dr_off/seed_1030/rec/check.json index 916c32c70377f02f8c5e2101e779fe301b7ddf67..99a7224db86a49eb1a9df794a75e8327db5a9eae 100644 --- a/N053/sawyer_umi/dr_off/seed_1030/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1030/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1030/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1030/rec/check.txt index 808634382127214b3945a43ed9eefa9d7acea376..809d0f7b18cf2fcacd3d1dab6f0669ac4f98e880 100644 --- a/N053/sawyer_umi/dr_off/seed_1030/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1030/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1030/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1490 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1490 at dataset fps 25 = 59.60 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1490 rows vs ceil(2979/2) = 1490 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3039, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 3e1cc78d4f45589ad8fab00077574ac1e696a180..8f297f04252fcba29f9ec949a9b2347292c5abc8 100644 --- a/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1490, "magicsim/control_steps": 2979, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1490, "magicsim/control_steps": 2979, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 02e68d7f6a43df7894d0d8fba33cf0989f642fb7..a0c4f367b1774007dc378ba7a066ebf76aa624c1 100644 --- a/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1030/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1031/rec/check.json b/N053/sawyer_umi/dr_off/seed_1031/rec/check.json index b244f459ddaa406b14664da9f03602807a776ff1..62e6c8ab13623c1e89ca13e7be4bffc9fb1d9b04 100644 --- a/N053/sawyer_umi/dr_off/seed_1031/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1031/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1031/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1031/rec/check.txt index 4c36e2799bc84ae22c56e83084ab275c968acc99..9aed64c34297fae8349332fc0484bb18a5f936ec 100644 --- a/N053/sawyer_umi/dr_off/seed_1031/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1031/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1031/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1671 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1671 at dataset fps 25 = 66.84 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1671 rows vs ceil(3342/2) = 1671 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3401, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index dabdaca412f5e268d7363d1e3bbb6c2216c90edd..bd204b7eccaf1707fe1c839ede1acc42a249e73f 100644 --- a/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1671, "magicsim/control_steps": 3342, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1671, "magicsim/control_steps": 3342, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index e4b5fae0b3abda16d31f69299b0ebd2dd24bf533..982bdf41fa46b71e809b3c384453a48597d4e447 100644 --- a/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1031/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1032/rec/check.json b/N053/sawyer_umi/dr_off/seed_1032/rec/check.json index 435afbbef7a51c5b3f1ff6393b9cef61f3b83b79..3151511e66c1e5fcc08d76caf214e92c3723c9a1 100644 --- a/N053/sawyer_umi/dr_off/seed_1032/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1032/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1032/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1032/rec/check.txt index 6a70eca8a1418c5694d006411a2421cf3ea74888..16181fa59ae86a45616cf8a0cb69cb872c1a4278 100644 --- a/N053/sawyer_umi/dr_off/seed_1032/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1032/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1032/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1496 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1496 at dataset fps 25 = 59.84 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1496 rows vs ceil(2991/2) = 1496 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3051, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index ce59567ba78af25cacb75b8dde3c0cc4065e9f9d..729a38fc38cbbb12d69ebda4507de7d44a0ea6f9 100644 --- a/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1496, "magicsim/control_steps": 2991, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1496, "magicsim/control_steps": 2991, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 214cfd58bd64efd93e5d9b256f6155b933916794..ed9df9e4ae255595632ea766cb95961f3417177c 100644 --- a/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1032/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1033/rec/check.json b/N053/sawyer_umi/dr_off/seed_1033/rec/check.json index 2c6ef66bf8dc6504ac87d14527b5a56b5d7d3cf2..84748711cecb877d19491bcebd094af844706db3 100644 --- a/N053/sawyer_umi/dr_off/seed_1033/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1033/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1033/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1033/rec/check.txt index 2a7c8938162c78a6d1fb715cdd8e864bd3317d5e..7c3f7a9602ca90b896d0a3ed3b06201f5e446acc 100644 --- a/N053/sawyer_umi/dr_off/seed_1033/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1033/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1033/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1633 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1633 at dataset fps 25 = 65.32 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1633 rows vs ceil(3266/2) = 1633 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3325, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 00ac0d3d34cb85cc8e88ddd627fbafa2ebea71a2..b058e4260b0cb09f21634f68aa252f6869a81857 100644 --- a/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1633, "magicsim/control_steps": 3266, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1633, "magicsim/control_steps": 3266, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index b8f31f013314ac416854670bdfba9f3ba6654158..683b2a59ec94dd9080a306370ab1da1a7b677d0e 100644 --- a/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1033/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1034/rec/check.json b/N053/sawyer_umi/dr_off/seed_1034/rec/check.json index 439aedc4721fff40a1cc87a1eefdd65cec580d70..ccdb0a841f14eab5b68a745ec38ebcf102944389 100644 --- a/N053/sawyer_umi/dr_off/seed_1034/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1034/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1034/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1034/rec/check.txt index c55b42bffae807ae9889026aca96eedae8895b62..e07156c20b924311df9c4ac4070ba098aa59bbcb 100644 --- a/N053/sawyer_umi/dr_off/seed_1034/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1034/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1034/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1645 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1645 at dataset fps 25 = 65.80 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1645 rows vs ceil(3290/2) = 1645 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3349, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index c9ab537fa115edb0665381bc9fb26e2f7b48ccf1..b773f24b0b2365dd6fd7f25f9b189a01dcfa86e4 100644 --- a/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1645, "magicsim/control_steps": 3290, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1645, "magicsim/control_steps": 3290, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index b0770061a8635fe126b2c4beb4f395ecc2041e98..da86d2e7966a4c4039da172aeb7fd864d3920926 100644 --- a/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1034/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1035/rec/check.json b/N053/sawyer_umi/dr_off/seed_1035/rec/check.json index de0a1c9a755e7bdf1ebeb8db27f0919162835e45..c9f662ac1c8789fc03cfca411d866ba1c7d189e8 100644 --- a/N053/sawyer_umi/dr_off/seed_1035/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1035/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1035/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1035/rec/check.txt index 3703a4c2662e2c2cdb713871b2e094c0465ac245..0162e182040f40158c8ce578ee3130b1864faf8d 100644 --- a/N053/sawyer_umi/dr_off/seed_1035/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1035/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1035/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1520 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1520 at dataset fps 25 = 60.80 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1520 rows vs ceil(3039/2) = 1520 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3099, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index eb675b6062466ae9a72f0b883e56e7d87ab752aa..43c59cb03cdd1fbdbbe94e905f321d31f82f6914 100644 --- a/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1520, "magicsim/control_steps": 3039, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1520, "magicsim/control_steps": 3039, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 200e7d7a38bd22ad00fc0dd67ee1e2f85448941a..604961dc2b9a6044d3be53b8f3735ee8ba7ed46a 100644 --- a/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1035/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1036/rec/check.json b/N053/sawyer_umi/dr_off/seed_1036/rec/check.json index 8b57687f15f22a9a3f4b13b6b83506a96e79b7d0..f9168ff006a2efd16b5be8144dcceb71d91b61f5 100644 --- a/N053/sawyer_umi/dr_off/seed_1036/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1036/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1036/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1036/rec/check.txt index 015460ce82d73ef6640d5fb6828ebe8930b05607..49180a375fca65649ffaf008fcfdc6d5417db8cc 100644 --- a/N053/sawyer_umi/dr_off/seed_1036/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1036/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1036/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1436 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1436 at dataset fps 25 = 57.44 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1436 rows vs ceil(2872/2) = 1436 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2931, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 86bfcdf374844a71da527041ce2769c0211a6917..139515a3b87794c9a188b8f659dee638064e101f 100644 --- a/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1436, "magicsim/control_steps": 2872, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1436, "magicsim/control_steps": 2872, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index f94aba01c2fa2e5e8e2eef31c47766c1e37509b3..65c5663b3b356a05bc9942ce4ad2d2efe1cf60ad 100644 --- a/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1036/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1037/rec/check.json b/N053/sawyer_umi/dr_off/seed_1037/rec/check.json index 2ba5612564df342d1a788833aa9a8b3908cdb851..3ea8aee47fd0d5799f7d0bc7d4af31011e2af8f2 100644 --- a/N053/sawyer_umi/dr_off/seed_1037/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1037/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1037/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1037/rec/check.txt index 598a26390f5ae8b707da6790f16ce783d42f9f48..88f4e5a44d2e1fe4ec7b67efd3e4b7c45803ebaf 100644 --- a/N053/sawyer_umi/dr_off/seed_1037/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1037/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1037_r1/out/rec/memorystage/ --- episode file episode_000000.parquet: 1468 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1468 at dataset fps 25 = 58.72 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1468 rows vs ceil(2935/2) = 1468 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2995, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 5168b152e35c35f2fca66fe6d25284b8c582f0a5..d41b06fab753a88782f74b5973f560b0c311fdb1 100644 --- a/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1468, "magicsim/control_steps": 2935, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1468, "magicsim/control_steps": 2935, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 7e02278a79c4138e9268a313a64fc90c3ad926e8..e953c3128225b8c282f74e38957d4ad938a9aa35 100644 --- a/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1037/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1038/rec/check.json b/N053/sawyer_umi/dr_off/seed_1038/rec/check.json index 3ad14e80a6b5979b51a49ca152e892009255b6a5..e287f5b07974f9eabe915a16f47fe925001b77c8 100644 --- a/N053/sawyer_umi/dr_off/seed_1038/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1038/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1038/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1038/rec/check.txt index 57cd7407d4b09585fdd448d10fca097af71ebfad..1d885354564ce36d64d26eedadcb0660583f3113 100644 --- a/N053/sawyer_umi/dr_off/seed_1038/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1038/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1038_r1/out/rec/memorystage/ --- episode file episode_000000.parquet: 1448 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1448 at dataset fps 25 = 57.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1448 rows vs ceil(2895/2) = 1448 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2955, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 2d28511b90871426db1e5f56dd7de931f6211318..53014273597257702d9ee89512aee1c06898f0fa 100644 --- a/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1448, "magicsim/control_steps": 2895, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1448, "magicsim/control_steps": 2895, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index b8ee5dfc435af9e5dbeb5cc9bc388cce19dd8d0e..2a33223ddbc84feddda9a6e6c4fe0c5ac2c89cad 100644 --- a/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1038/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1039/rec/check.json b/N053/sawyer_umi/dr_off/seed_1039/rec/check.json index 9def2f47b575c777ddf8ba772aa1320107115add..cb13fa29d6a3b0cf81ce6c66cec24dbc70c497e1 100644 --- a/N053/sawyer_umi/dr_off/seed_1039/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1039/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1039/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1039/rec/check.txt index df947ef6b6df1b628ae5549b124dfcf5eedfcc32..57025b0e02b20fbef34eff57bb6c8fd00e120e67 100644 --- a/N053/sawyer_umi/dr_off/seed_1039/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1039/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1039/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1649 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1649 at dataset fps 25 = 65.96 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1649 rows vs ceil(3297/2) = 1649 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3357, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 719c670c4785f08388f94a2d9c5741c3a3f21824..98f74a1b14e20f366d43c46eca23007f16b3fd5a 100644 --- a/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1649, "magicsim/control_steps": 3297, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1649, "magicsim/control_steps": 3297, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index b8d4808f6993c78f02d00420ea2d346de06c38f4..37fa8fd9428b06969149d57deb606445d049a372 100644 --- a/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1039/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1040/rec/check.json b/N053/sawyer_umi/dr_off/seed_1040/rec/check.json index 19ee3d2fd2f1d9da50d26e15da906dfd5450b15e..e259a28ddcf73e95f1b1f53861db816ecbf31945 100644 --- a/N053/sawyer_umi/dr_off/seed_1040/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1040/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1040/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1040/rec/check.txt index f2fbaadf2892b68e9ff43ec6cb1aa7ad7e08a13b..91f0da68c5f6891bf97b1096dc653af50aaef935 100644 --- a/N053/sawyer_umi/dr_off/seed_1040/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1040/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1040/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1519 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1519 at dataset fps 25 = 60.76 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1519 rows vs ceil(3037/2) = 1519 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3097, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index b25b555913ff071cf534d0fdb28129404f0ab437..2cb5f94301c9bc35a422e22af5040fcb8e6e7b35 100644 --- a/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1519, "magicsim/control_steps": 3037, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1519, "magicsim/control_steps": 3037, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 6f5da685167294a284bc21611405b931a339560a..116a341d20bf2733a755db72ba53bc5985e5e7c0 100644 --- a/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1040/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1041/rec/check.json b/N053/sawyer_umi/dr_off/seed_1041/rec/check.json index e81f46c00c8324104c7c593b1b6568cc49c36d9b..c78693eefd4955c9191740b6bce9a1a7d6c163eb 100644 --- a/N053/sawyer_umi/dr_off/seed_1041/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1041/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1041/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1041/rec/check.txt index 0da76585c8f03ee69927863133c26708794aa766..e424bc6914d30387a9bc10f5ccfcdfdea41cdee0 100644 --- a/N053/sawyer_umi/dr_off/seed_1041/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1041/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1041/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1645 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1645 at dataset fps 25 = 65.80 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1645 rows vs ceil(3290/2) = 1645 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3349, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index ae53b7ada54313fa09fb484a258b151540fe1d53..859fcdb0dca877fafe04637e89db9c1310b97b2e 100644 --- a/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1645, "magicsim/control_steps": 3290, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1645, "magicsim/control_steps": 3290, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index da1842ff97c5dc628ba454e3765fcdfae6ed50a5..3f4a7c4a00ed18feb608266c6d01ab9f4a724545 100644 --- a/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1041/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1042/rec/check.json b/N053/sawyer_umi/dr_off/seed_1042/rec/check.json index 0014d07cf2d056a3235b6adbf08bc04899b754d2..0f95489e4f576dacd0fffd3b7a9ac2fb23f22520 100644 --- a/N053/sawyer_umi/dr_off/seed_1042/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1042/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1042/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1042/rec/check.txt index ea7d6c34944eee17a3546c99dcb7d398b97a9372..2add287947cb10f66cbb9bc669e6ae64fb577df7 100644 --- a/N053/sawyer_umi/dr_off/seed_1042/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1042/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1042/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1608 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1608 at dataset fps 25 = 64.32 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1608 rows vs ceil(3216/2) = 1608 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3275, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index bbb5169336cf605ae44bc3c2d8321ff698394966..7ee94449880125cdd0761bd5f378a308d7d637ab 100644 --- a/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1608, "magicsim/control_steps": 3216, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1608, "magicsim/control_steps": 3216, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 12351ca5197f22dbcd8fd1ee96f7273b25d4e4d2..52f9c09d0dab1267efc1377333c0ebb6359214bb 100644 --- a/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1042/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1043/rec/check.json b/N053/sawyer_umi/dr_off/seed_1043/rec/check.json index baba97c7df14439fddb22b897f45c64627b57147..e26854eb083e273354778d6c5f5c142a1b28cff4 100644 --- a/N053/sawyer_umi/dr_off/seed_1043/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1043/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1043/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1043/rec/check.txt index 24301108ea0ff9fcad23cd6840662ea6dac72e8c..2c42b13c9c2085d79433530d2446adc423f9662d 100644 --- a/N053/sawyer_umi/dr_off/seed_1043/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1043/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1043/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1666 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1666 at dataset fps 25 = 66.64 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1666 rows vs ceil(3332/2) = 1666 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3391, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 0f1f1d8a500eb3ef42c0bdcea2fba5ebda9d3b7d..9b9d3d4c94ca2c2c7db13a3488aba0d24e9f712e 100644 --- a/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1666, "magicsim/control_steps": 3332, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1666, "magicsim/control_steps": 3332, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 71bbce2f93f3a8a7f61eb6142bb051354566931c..1baed929948d29852da2cf9ee1993477fb534939 100644 --- a/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1043/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1044/rec/check.json b/N053/sawyer_umi/dr_off/seed_1044/rec/check.json index a4c25a2f761e21201b455ac862294faf97e6e6bc..959c643607d862e89ade307f5266442906e62975 100644 --- a/N053/sawyer_umi/dr_off/seed_1044/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1044/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1044/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1044/rec/check.txt index adeb78fb26d68dc359bbf18d321799dd552f0fd2..37f0c7431292f57ed1818391047306fe9966ab24 100644 --- a/N053/sawyer_umi/dr_off/seed_1044/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1044/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1044/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1487 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1487 at dataset fps 25 = 59.48 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1487 rows vs ceil(2974/2) = 1487 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3033, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index de89c7afa25a69641a0acfb0cc1ea54eaa704177..69b28a6ca697296656d9780d1c65b7fc66975c07 100644 --- a/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1487, "magicsim/control_steps": 2974, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1487, "magicsim/control_steps": 2974, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index d62fb8fc431b259ed3df4642cea674a5887c96c2..6ec08e59bc951cb124606382cc872665247b63ca 100644 --- a/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1044/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1045/rec/check.json b/N053/sawyer_umi/dr_off/seed_1045/rec/check.json index c760b0a60d8e8f1d856b243ba10ead92ee2cdbe7..c285209dc7587ce5c31787d4e735c07d1ea1ce48 100644 --- a/N053/sawyer_umi/dr_off/seed_1045/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1045/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1045/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1045/rec/check.txt index 3b1fa3b7a1b39a65eadeb71642b4ee928dfb7a96..c33e9cb851ee00b9371953cc98c9a60cf5bce3ba 100644 --- a/N053/sawyer_umi/dr_off/seed_1045/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1045/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1045/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1705 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1705 at dataset fps 25 = 68.20 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1705 rows vs ceil(3409/2) = 1705 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3469, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 75fa002265c6e0234086fcf8c1fd23289bde14df..2ddd3cb781e347138c4550f8e33ad6bd6377f9cd 100644 --- a/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1705, "magicsim/control_steps": 3409, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1705, "magicsim/control_steps": 3409, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index e4d725a0da0fdc8bc51c550a447fe468d9e5d443..921ff560a98f621941953670f3be0777038b72db 100644 --- a/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1045/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1046/rec/check.json b/N053/sawyer_umi/dr_off/seed_1046/rec/check.json index 5ce988e1e2768313824a0fb20bde420a060f7eaa..90dc91dfd258bdd648af7e20a4f56d73c02f1f85 100644 --- a/N053/sawyer_umi/dr_off/seed_1046/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1046/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1046/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1046/rec/check.txt index 65ca62691080bdc377c6cc0f732c789cf982cc79..784fa6f2287c9d71245f8f16b251eb9e7eebfd13 100644 --- a/N053/sawyer_umi/dr_off/seed_1046/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1046/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1046/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1487 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1487 at dataset fps 25 = 59.48 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1487 rows vs ceil(2973/2) = 1487 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3033, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index f8dadb9071fece85d2229f61616e626ea62b57aa..d26929e34c760a7aaa2d2a0cf64b522fed517759 100644 --- a/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1487, "magicsim/control_steps": 2973, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1487, "magicsim/control_steps": 2973, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index fc3199a3db5201a14101b2ca2631c048e43e4f4e..cf50a71f9f682649163efdec23f7da351c93d20b 100644 --- a/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1046/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1047/rec/check.json b/N053/sawyer_umi/dr_off/seed_1047/rec/check.json index 82a11fa7b35cb40e478ab20290b83a6c2f840993..201bf70fb5e5b1811904db137c074cbb3c75320e 100644 --- a/N053/sawyer_umi/dr_off/seed_1047/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1047/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1047/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1047/rec/check.txt index bbdfe217df53ec3b374d7ef3adc32d8d609c59b0..9e535b7a94037bdac69cc6233ea21c1e5792f400 100644 --- a/N053/sawyer_umi/dr_off/seed_1047/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1047/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1047/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1534 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1534 at dataset fps 25 = 61.36 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1534 rows vs ceil(3068/2) = 1534 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3127, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 2a3ef418c58d8d29f2330b09e11a086ff4b8be4d..eb81f1dae375fae32b7fa92e6f65d224e96c8c77 100644 --- a/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1534, "magicsim/control_steps": 3068, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1534, "magicsim/control_steps": 3068, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index c98555a1652810ce1fbc0aea2e2653a53ca9c63d..af44b59f69cb93566729cbb25c567f9490f14669 100644 --- a/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1047/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1048/rec/check.json b/N053/sawyer_umi/dr_off/seed_1048/rec/check.json index 9c9f8bb4632c83361cbffa257d90cbf1f52b0b3a..a46b428b56457169bcd613a79be874aaaa1c8e38 100644 --- a/N053/sawyer_umi/dr_off/seed_1048/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1048/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1048/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1048/rec/check.txt index 98b7dc6a727afe82ce709a5f5690bda6001b0603..a157bc33e6432cacdaa6b2564f5d0b1e2ee9ffed 100644 --- a/N053/sawyer_umi/dr_off/seed_1048/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1048/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1048/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1648 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1648 at dataset fps 25 = 65.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1648 rows vs ceil(3296/2) = 1648 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3355, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index ae0ff467eea6e59a8774e52246fe42aa271ebe2d..6947f3123bcdeb33f85c579e0bb3e63398737924 100644 --- a/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1648, "magicsim/control_steps": 3296, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1648, "magicsim/control_steps": 3296, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 692f3707170e6e418306feae79f68ee5bd99dabb..4e462e87b2ccfc7f09ff1ccec25e0c4ca08c4885 100644 --- a/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1048/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1049/rec/check.json b/N053/sawyer_umi/dr_off/seed_1049/rec/check.json index 7192be56b6b3659adc3a48352116c175ff01cf26..c981fc500c6fc30f84f6697f3b85ef15aef6ca08 100644 --- a/N053/sawyer_umi/dr_off/seed_1049/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1049/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1049/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1049/rec/check.txt index cab082defaec3ecdcfb94a55ac2c7fccc8ce7a15..b4d1058e299bf4881dc5ac7b573ae9422165879f 100644 --- a/N053/sawyer_umi/dr_off/seed_1049/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1049/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1049/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1666 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1666 at dataset fps 25 = 66.64 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1666 rows vs ceil(3332/2) = 1666 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3391, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index f093438513b93d0b7fb56b65b661e1d435c8037b..9fcfe09250a6a8be76f9f11a8379d8a008610e01 100644 --- a/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1666, "magicsim/control_steps": 3332, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1666, "magicsim/control_steps": 3332, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index edec5f0c8265c7e0abfb455ea4cedc17685f5ba9..de4777e139d45ba5cbe42b379ca5fd05638b52d9 100644 --- a/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1049/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1050/rec/check.json b/N053/sawyer_umi/dr_off/seed_1050/rec/check.json index 79cd906917b4d02bfbb89a0db2883f69120193cd..b917bc5bf93f056dd61101d5dbea912d110f9149 100644 --- a/N053/sawyer_umi/dr_off/seed_1050/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1050/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1050/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1050/rec/check.txt index 3f83b42a500e48e654892480032388254e977c76..9773347275a86b5e5c550cb6b9d98bdbd593b00e 100644 --- a/N053/sawyer_umi/dr_off/seed_1050/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1050/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1050/out/rec/memorystage/000 --- episode file episode_000000.parquet: 1448 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1448 at dataset fps 25 = 57.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1448 rows vs ceil(2896/2) = 1448 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2955, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index fa9e034cff9b4a9b9ff5024e841e888887460d59..f96242d22ceda83274e94fc85bec7ada8232d517 100644 --- a/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1448, "magicsim/control_steps": 2896, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1448, "magicsim/control_steps": 2896, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 9df0b20ac86ae07a9cf6ef883bf98b1d368792b3..083153e83b639f06b581d84a6a30bb17f4d81584 100644 --- a/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1050/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_off/seed_1051/rec/check.json b/N053/sawyer_umi/dr_off/seed_1051/rec/check.json index 75464c0551f6b3f743530bdf8dde10bafe9732ab..87bd8a34ae08304b2c1ea293e36fadb5228179bb 100644 --- a/N053/sawyer_umi/dr_off/seed_1051/rec/check.json +++ b/N053/sawyer_umi/dr_off/seed_1051/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_off/seed_1051/rec/check.txt b/N053/sawyer_umi/dr_off/seed_1051/rec/check.txt index d1c07761fe83e933e583ac964fd56e0f23474435..787d22db4623675a8dbf4d96fddf87ec85121fbb 100644 --- a/N053/sawyer_umi/dr_off/seed_1051/rec/check.txt +++ b/N053/sawyer_umi/dr_off/seed_1051/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_off_1051_r1/out/rec/memorystage/ --- episode file episode_000000.parquet: 1448 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1448 at dataset fps 25 = 57.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1448 rows vs ceil(2896/2) = 1448 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2955, step differences [2] diff --git a/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 3f4371fea1ccd1b45a72abeb1cd60ef324bc6f2a..7d5af4d78e93cf204daf02e4f0c36617dbca312b 100644 --- a/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1448, "magicsim/control_steps": 2896, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1448, "magicsim/control_steps": 2896, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 571c9464bd52d725afa3768c7ba5a87087f21fab..7202f959ff1164bceb54be24094181190d9b96d6 100644 --- a/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_off/seed_1051/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2001/rec/check.json b/N053/sawyer_umi/dr_on/seed_2001/rec/check.json index 5eebeff8f0718492accd8c39e8ed37f3974f9ae0..edbadced3fab9339b6e543917157d317cbcbd267 100644 --- a/N053/sawyer_umi/dr_on/seed_2001/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2001/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2001/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2001/rec/check.txt index a42683687c6b2ada620b243f597c7f39c43511f3..517a466fd22022b90bb285c0c1dde44c972fcdbd 100644 --- a/N053/sawyer_umi/dr_on/seed_2001/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2001/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2001/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1513 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1513 at dataset fps 25 = 60.52 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1513 rows vs ceil(3026/2) = 1513 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3085, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 64c81896772d03faaadb244ce45c819a2cc745d1..f731105997df556f47fdccb9cc93cda7897aa5c7 100644 --- a/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1513, "magicsim/control_steps": 3026, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1513, "magicsim/control_steps": 3026, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index bdca4e811330394be8f680ec288ba3e59aa1edd4..8ae4fc3a088e6934c1e60ca1097a4785b43a05e2 100644 --- a/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2001/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2002/rec/check.json b/N053/sawyer_umi/dr_on/seed_2002/rec/check.json index 5eb495c3954e723069d2fbac36c09b41b3e45072..d5cea62c7f38d41cbcb0f0fe40caebe847508b9e 100644 --- a/N053/sawyer_umi/dr_on/seed_2002/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2002/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2002/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2002/rec/check.txt index e1bc5b267ae60a93a6bbd7ffac68a51034de3327..58a14ec1ef762c3bfe00383d9ffceb4bd9895af3 100644 --- a/N053/sawyer_umi/dr_on/seed_2002/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2002/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2002/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1638 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1638 at dataset fps 25 = 65.52 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1638 rows vs ceil(3275/2) = 1638 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3335, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 636218d6f9d0040058ed5888550de7738f4e734d..c16f8cc51c1c2f231f8222523ff15c0ebdc12a4e 100644 --- a/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1638, "magicsim/control_steps": 3275, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1638, "magicsim/control_steps": 3275, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the right of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 8cdc33f0f76ac8498865f97f0cdd8aa2ac6b5561..4f10e1b0b77ef8df8405e3f0347c9178858bdaa3 100644 --- a/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -529,7 +529,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2002/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2003/rec/check.json b/N053/sawyer_umi/dr_on/seed_2003/rec/check.json index 3939b15561295b95e60d559605b5c7cd1ece05fc..8be7ff85aa264526be515a9776bd2c2e0d577629 100644 --- a/N053/sawyer_umi/dr_on/seed_2003/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2003/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2003/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2003/rec/check.txt index dc6e599fc3f7fece4c2d2e4e658d56bc23e63fd4..44bfe9fcba3696db06f2dddcc27857105bdca03f 100644 --- a/N053/sawyer_umi/dr_on/seed_2003/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2003/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2003/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1661 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1661 at dataset fps 25 = 66.44 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1661 rows vs ceil(3321/2) = 1661 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3381, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 80f2ff6285ef503fa4a5308d49112d9e2a229484..61cfe6fcdf3eb40608d438cfba4c54413dc68c6e 100644 --- a/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1661, "magicsim/control_steps": 3321, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1661, "magicsim/control_steps": 3321, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 6637e18a982719705801abf5ffb9494ae4360eaf..5f1ae3a49ae62192b212e1978ef71c44a4434579 100644 --- a/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2003/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2004/rec/check.json b/N053/sawyer_umi/dr_on/seed_2004/rec/check.json index c8f2b77415bd7976f1f92a2b27a7ca2d42fd5770..ab49149b269eb91416ab96c24e2ec0586e891f51 100644 --- a/N053/sawyer_umi/dr_on/seed_2004/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2004/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2004/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2004/rec/check.txt index b9a785d39e26f131e7161787843ad979d229a89f..a893787bf2764c7ff3812c5eb9f38de04debfe31 100644 --- a/N053/sawyer_umi/dr_on/seed_2004/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2004/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2004/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1673 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1673 at dataset fps 25 = 66.92 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1673 rows vs ceil(3346/2) = 1673 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3405, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 0290802084cede3ac98467739c28304034e07dd6..533c6bd358a4808fccad26a13b70daeb200fb64c 100644 --- a/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1673, "magicsim/control_steps": 3346, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1673, "magicsim/control_steps": 3346, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index 204a698d371cebfa646d895febaa057f0156d449..0f2a2dd69ece545e84fdeb628fd70a91d768aa3c 100644 --- a/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2004/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2005/rec/check.json b/N053/sawyer_umi/dr_on/seed_2005/rec/check.json index 36a2aec8429c0da007976e3870d65100e8bc4158..1db2abf516d610566cce5abcde99c7960d0144e0 100644 --- a/N053/sawyer_umi/dr_on/seed_2005/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2005/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2005/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2005/rec/check.txt index 5034d6a17419f8ce930d1986be9cd0246dce53b8..5a9d1a9741d2e28dfa95d9f38d36ab7d9f9d8197 100644 --- a/N053/sawyer_umi/dr_on/seed_2005/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2005/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2005/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1463 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1463 at dataset fps 25 = 58.52 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1463 rows vs ceil(2926/2) = 1463 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..2985, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 4f3e80aa3429f4b52c5a23c08973ed79b27259c5..95e3b5ce4a4c65620709f64bdf2bcc28200c5fd6 100644 --- a/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1463, "magicsim/control_steps": 2926, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1463, "magicsim/control_steps": 2926, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is in front of the rack. The white st booth r is in front of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index d662896eeb322d344871d28ca898a54e226e00ba..0c6fb11ab317b66e5f58dfee5b60772cc805652d 100644 --- a/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2005/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2006/rec/check.json b/N053/sawyer_umi/dr_on/seed_2006/rec/check.json index e956cab7c5b2e236b4e0c37054cb5c024d493270..b2e1397a3388fc5368f2a3c809f6bfa9fec36399 100644 --- a/N053/sawyer_umi/dr_on/seed_2006/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2006/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2006/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2006/rec/check.txt index 0a96164cce53ed7030dc934d76f6f4c609fdb35c..1c791c576abdc58c0c04b614ac173284277928e3 100644 --- a/N053/sawyer_umi/dr_on/seed_2006/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2006/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2006/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1531 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1531 at dataset fps 25 = 61.24 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1531 rows vs ceil(3062/2) = 1531 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3121, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 3a0c1c31c024f57367036c53016ba30b244cba60..567d7e4541e760de238c99032e929c2497d4c9cb 100644 --- a/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1531, "magicsim/control_steps": 3062, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1531, "magicsim/control_steps": 3062, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index dc5adf9443cbc7efacabc69d9ebffd59bf0580d0..fe81229ddc9b29cae38af5d9cb6495b64c986671 100644 --- a/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2006/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2007/rec/check.json b/N053/sawyer_umi/dr_on/seed_2007/rec/check.json index b306516ca7021119ba41ca3ea5c12e13c70c3c8b..52f297df4291ad3dcb81c25e2cb9bdb17ca696b4 100644 --- a/N053/sawyer_umi/dr_on/seed_2007/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2007/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2007/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2007/rec/check.txt index 9ce007207855a63a970eff339aa1f4d988171d94..c8af7a974a207232024f18eaeffdb8a0e2ad58f0 100644 --- a/N053/sawyer_umi/dr_on/seed_2007/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2007/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2007/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1499 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1499 at dataset fps 25 = 59.96 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1499 rows vs ceil(2997/2) = 1499 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3057, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 976d4661c934a68d5ee7ea3e750204507163d655..e5cbe6d4ce15782341a4845521328f3cc83eee64 100644 --- a/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1499, "magicsim/control_steps": 2997, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1499, "magicsim/control_steps": 2997, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is on top of the rack. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the right of the rack. The white st booth r is to the left of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The gray pad is to the left of the pen s1. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index f406f639083fdb2da4c6aed1023ef3b8e51c32f3..f02a98a1b606d49f4b947d838f88c0fd66d3a863 100644 --- a/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -534,7 +534,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2007/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."} diff --git a/N053/sawyer_umi/dr_on/seed_2008/rec/check.json b/N053/sawyer_umi/dr_on/seed_2008/rec/check.json index 22e3d203195faa727470f52925a4a472505cc508..e947dd4b9a5288a074cafd61c17bd70cd67f93f1 100644 --- a/N053/sawyer_umi/dr_on/seed_2008/rec/check.json +++ b/N053/sawyer_umi/dr_on/seed_2008/rec/check.json @@ -23,7 +23,7 @@ }, "task text = MobileBench instruction": { "pass": true, - "detail": "task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'" + "detail": "task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'" }, "rows = recorded control steps / 2 (= episode length at 25 Hz)": { "pass": true, diff --git a/N053/sawyer_umi/dr_on/seed_2008/rec/check.txt b/N053/sawyer_umi/dr_on/seed_2008/rec/check.txt index 47d43f2361cdee179c1758cab453a85d071d700e..d76f7d0c99cb5262d662fdf64153c06e1e5f6748 100644 --- a/N053/sawyer_umi/dr_on/seed_2008/rec/check.txt +++ b/N053/sawyer_umi/dr_on/seed_2008/rec/check.txt @@ -7,7 +7,7 @@ dataset root /mnt/scratch/sz850_atomic/n053_saw_on_2008/out/rec/memorystage/000/ --- episode file episode_000000.parquet: 1614 rows, columns ['action', 'action_is_hold', 'observation.state', 'observation.joint_vel', 'observation.eef_pose', 'observation.environment_state', 'action.command', 'action.applied', 'narration', 'collect_info', 'extra.reward', 'extra.action_info', 'next.truncated', 'sim_step_idx', 'timestamp', 'frame_index', 'episode_index', 'index', 'task_index', 'next.done'] [PASS] observation.state shape: parquet width 15, info shape [15], 15 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'right_j0', 'head_pan', 'right_j1'] [PASS] action shape: parquet width 11, info shape [11], 11 names first: ['dummy_base_prismatic_x_joint', 'dummy_base_prismatic_y_joint', 'dummy_base_revolute_z_joint', 'frame0.pos_x', 'frame0.pos_y', 'frame0.pos_z'] -[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.'; expected 'Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.' +[PASS] task text = MobileBench instruction: task_index 0: 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.'; expected 'Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.' rows 1614 at dataset fps 25 = 64.56 s [PASS] rows = recorded control steps / 2 (= episode length at 25 Hz): 1614 rows vs ceil(3228/2) = 1614 [PASS] rows are every 2nd control step (sim_step_idx): sim_step_idx 61..3287, step differences [2] diff --git a/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl b/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl index 4a9648496c926344f6e62f1780437c7be398c042..ca48682f89de6e954edc4f8d676c41c75366369b 100644 --- a/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/episodes.jsonl @@ -1 +1 @@ -{"episode_index": 0, "tasks": ["Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."], "length": 1614, "magicsim/control_steps": 3228, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} +{"episode_index": 0, "tasks": ["Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."], "length": 1614, "magicsim/control_steps": 3228, "magicsim/subsample_every": 2, "magicsim/env_id": 0, "magicsim/trajectory_id": 0, "magicsim/scene_caption": "A Ridgeback-Sawyer mobile manipulator with a UMI gripper is positioned at the workspace, facing the rack. There are a large white st wall, a large gray st table a, a large gray st table b, a large white st booth l, a large white st booth r, a medium rack, a small pen s0, a small pen s1, a large gray pad, a large red tag red and a large blue tag blue. The white st wall is to the left of the rack. The gray st table a is behind the rack. The gray st table b is to the right of the pen s1. The white st booth l is on top of the rack. The white st booth l is to the left of the rack. The white st booth r is to the right of the rack. The rack is in front of the gray st table a. The pen s0 is to the right of the red tag red. The pen s1 is to the right of the blue tag blue. The red tag red is to the left of the pen s0. The blue tag blue is to the left of the pen s1. The scene is viewed from left, at table height."} diff --git a/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json b/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json index fa7302f4825da1232de440492da66f2a3631c637..ed2dd25b684c30c2ee414a03bb80edc5016ba2f1 100644 --- a/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json +++ b/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/extra/episode_000000.json @@ -524,7 +524,7 @@ }, "task": { "task_type": "MemoryStage", - "instruction": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it.", + "instruction": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad.", "template_id": -1, "slots": { "obj": 1.5707963267948966, diff --git a/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl b/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl index 577f78948573fbc166b2f4f9635e6ef0aae713bf..601389552aa6182826e3a2e4fe872a274827ef9c 100644 --- a/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl +++ b/N053/sawyer_umi/dr_on/seed_2008/rec/memorystage/000/lerobot_v21/meta/tasks.jsonl @@ -1 +1 @@ -{"task_index": 0, "task": "Watch which pen touches the red zone, then go around the partition and touch the target pad with the pen that did not touch it."} +{"task_index": 0, "task": "Watch which pen touches the red panel, then go around the partition and put the pen that did not touch it on the target pad."}