Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    ArrowInvalid
Message:      Schema at index 20 was different: 
task: string
quantize: int64
grasp: int64
success: int64
grasp_rate: double
success_rate: double
num_episodes: int64
avg_inference_time: double
avg_duration_sec: double
avg_ticks: double
avg_lag_l2: double
vs
task: string
grasp: int64
success: int64
grasp_rate: double
success_rate: double
num_episodes: int64
avg_inference_time: double
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1869, in _prepare_split_single
                  num_examples, num_bytes = writer.finalize()
                                            ~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 781, in finalize
                  self.write_rows_on_file()
                  ~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 662, in write_rows_on_file
                  table = pa.concat_tables(self.current_rows)
                File "pyarrow/table.pxi", line 6321, in pyarrow.lib.concat_tables
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                  return check_status(status)
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
                  raise convert_status(status)
              pyarrow.lib.ArrowInvalid: Schema at index 20 was different: 
              task: string
              quantize: int64
              grasp: int64
              success: int64
              grasp_rate: double
              success_rate: double
              num_episodes: int64
              avg_inference_time: double
              avg_duration_sec: double
              avg_ticks: double
              avg_lag_l2: double
              vs
              task: string
              grasp: int64
              success: int64
              grasp_rate: double
              success_rate: double
              num_episodes: int64
              avg_inference_time: double
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1880, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

task
string
quantize
int64
grasp
int64
success
int64
grasp_rate
float64
success_rate
float64
num_episodes
int64
avg_inference_time
float64
avg_duration_sec
float64
avg_ticks
float64
avg_lag_l2
float64
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
20
20
62.5
62.5
32
0.0951
14.519
136.7
0.0871
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
0
0
0
0
1
0.1134
null
null
null
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
4
4
80
80
5
0.0866
14.841
141.2
0.0881
Pick up the blue cup and put it into the plate
1
1
1
100
100
1
0.1122
8.732
80
0.0779
Pick up the doll and put it into the plate
1
5
5
83.333
83.333
6
0.0972
6.654
60.8
0.0887
Pick up the green cube and put it into the plate
1
8
8
72.727
72.727
11
0.0843
7.401
68.1
0.0732
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
3
3
100
100
3
0.0923
22.413
204.7
0.0523
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
2
2
100
100
2
0.0819
20.305
192
0.0585
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
2
2
100
100
2
0.1131
17.066
156.5
0.0578
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
2
2
100
100
2
0.0887
20.694
193.5
0.0573
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
3
3
100
100
3
0.092
20.472
196
0.0553
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
26
26
81.25
81.25
32
0.0979
19.39
181.1
0.0567
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
1
1
100
100
1
0.1155
25.974
237
0.0576
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
2
2
100
100
2
0.1005
22.464
208.5
0.0544
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
9
9
100
100
9
0.0889
25.019
243
0.0647
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
2
1
1
50
50
2
0.1106
13.434
126
0.0831
Pick up the blue cup and put it into the plate
1
3
3
50
50
6
0.0965
13.757
135
0.0595
Pick up the doll and put it into the plate
1
15
15
78.947
78.947
19
0.1024
10.428
100.9
0.0585
Pick up the green cube and put it into the plate
1
17
17
100
100
17
0.1084
12.206
117.9
0.0529
Pick up the blue cup and put it into the plate
2
9
9
81.818
81.818
11
0.0898
8.276
78
0.0654
Pick up the doll and put it into the plate
2
5
5
50
50
10
0.1126
6.699
63.6
0.0763
Pick up the green cube and put it into the plate
2
6
6
54.545
54.545
11
0.1154
7.52
71
0.0667
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
27
27
87.097
87.097
31
0
35.204
341.3
0.0528
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
2
0
0
0
0
1
0.1053
null
null
0.0642
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
0
0
0
0
2
0.0791
null
null
0.0673
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
3
3
75
75
4
0.1259
19.467
188
0.0678
Pick up the blue cup and put it into the plate
1
5
5
45.455
45.455
11
0.1093
9.539
92.6
0.0646
Pick up the doll and put it into the plate
1
8
8
80
80
10
0.0892
7.105
68
0.0729
Pick up the green cube and put it into the plate
1
7
7
63.636
63.636
11
0.1449
8.41
80.1
0.0617
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
4
4
33.333
33.333
12
0.1252
24.276
236
0.0565
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
1
1
1
25
25
4
0.0932
14.113
140
0.0519
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
0
0
0
0
1
0.1134
null
null
0.0616
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
1
1
100
100
1
0.098
42
409
0.0459
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
2
11
11
73.333
73.333
15
0.0956
24.751
235.2
0.0603
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
3
3
3
50
50
6
0.1054
19.646
182.7
0.0718
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
4
1
1
50
50
2
0.09
20.453
190
0.0598
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
1
1
100
100
1
0.1175
21.985
212
0.0523
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
0
0
0
0
1
0.1263
null
null
0.0619
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
0
0
0
0
1
0.2471
null
null
0.0562
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
0
0
0
0
3
0.2014
null
null
0.0498
Pick up the blue cup and put it into the plate
1
0
0
0
0
1
0.2206
null
null
0.069
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
1
0
0
0
0
3
0.0945
null
null
0.0545
Pick up the blue cup and put it into the plate
1
2
2
100
100
2
0.1178
6.703
64.5
0.0758
Pick up the doll and put it into the plate
1
8
8
100
100
8
0.1039
5.628
54.4
0.0791
Pick up the green cube and put it into the plate
1
0
0
0
0
1
0.0616
null
null
0.0705
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
11
11
55
55
20
0.0761
15.787
154.4
0.0665
Pick up the blue cup and put it into the plate
1
4
4
100
100
4
0.0718
12.874
127
0.0572
Pick up the doll and put it into the plate
1
13
13
100
100
13
0.0985
10.394
100.5
0.0616
Pick up the green cube and put it into the plate
1
9
9
75
75
12
0.0893
11.805
115.2
0.0577
Pick up the blue cup and put it into the plate
2
11
11
84.615
84.615
13
0.0944
8.315
79.1
0.072
Pick up the doll and put it into the plate
2
2
2
100
100
2
0.0811
5.844
56.5
0.0777
Pick up the green cube and put it into the plate
2
12
12
100
100
12
0.0841
6.381
61.5
0.0792
Pick up the blue cup and put it into the plate
3
9
9
75
75
12
0.0978
6.322
58.8
0.086
Pick up the doll and put it into the plate
3
10
10
83.333
83.333
12
0.0881
5.026
47
0.091
Pick up the green cube and put it into the plate
3
14
14
100
100
14
0.0916
5.542
51.8
0.0907
Pick up the blue cup and put it into the plate
4
2
2
100
100
2
0.1106
5.238
47
0.0934
Pick up the doll and put it into the plate
4
10
10
83.333
83.333
12
0.0877
4.452
41.1
0.1038
Pick up the green cube and put it into the plate
4
10
10
83.333
83.333
12
0.1005
5.071
45.9
0.1112
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
1
1
11.111
11.111
9
0.0758
23.208
226
0.0592
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
1
1
1
25
25
4
0.0773
14.025
138
0.0548
Open the air fryer basket, put the food from the plate into the basket, and close the basket.
null
18
18
56.25
56.25
32
0.0797
null
null
null
Put the doll and the green cube in the left red cup, and put the mint cube in the right red cup.
null
19
19
61.29
61.29
31
0.0798
null
null
null
Pick up the blue cup and put it into the plate
null
5
5
45.455
45.455
11
0.0832
null
null
null
Pick up the doll and put it into the plate
null
10
10
100
100
10
0.0801
null
null
null
Pick up the green cube and put it into the plate
null
6
6
54.545
54.545
11
0.0801
null
null
null
Pick up the blue cup and put it into the plate
null
0
0
0
0
11
0.0856
null
null
null
Pick up the doll and put it into the plate
null
9
9
90
90
10
0.0835
null
null
null
Pick up the green cube and put it into the plate
null
8
8
72.727
72.727
11
0.0817
null
null
null
Pick up the blue cup and put it into the plate
null
0
0
0
0
11
0.0851
null
null
null
Pick up the doll and put it into the plate
null
10
10
83.333
83.333
12
0.0846
null
null
null
Pick up the green cube and put it into the plate
null
8
8
72.727
72.727
11
0.0845
null
null
null
Stack the left blue cup on the yellow cup in the middle, then stack the right blue cup on the yellow cup in the middle
null
11
11
34.375
34.375
32
0.0797
null
null
null

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

What gets logged

Tree: logs/base/<model>/<variant>/<model>/<task>/{results,videos}/ where <variant> is q1..q4. (logs/sail, logs/demospeedup have the same shape, one root per runner script.) Cross-variant summaries live in ../result/<variant>.csv, not here.

Join key: episode_id. Same value in episodes.tsv and tracking.tsv; the video is videos/{episode_id + 1:03d}.mp4 (episode_id 12 → 013.mp4).

Two dt bases appear throughout. Unprefixed columns use the measured interval between consecutive commands — what the arm actually faced, and the numbers to quote. nominal_* recompute the same thing as if every interval had been exactly 1/10 s; the gap between them is loop timing, not quantization. Diagnostic only.


results/episodes.tsv — one row per episode

핵심

열 뜻
episode_id 조인 키 (영상은 +1번)
success 1.0 성공 / 0.5 잡기만 / 0.0 실패
ticks, duration_sec 명령 횟수, 걸린 시간 — 속도 이득의 분자·분모
mean_lag_l2 명령과 실측 관절의 평균 오차 (rad). 크면 로봇이 명령을 못 따라간 것
torque_saturated_ticks 정격 토크를 넘긴 tick 수
max_torque_frac_of_rated 그 에피소드 최대 |τ|/정격. > 1 이면 명령대로 못 감
per_joint_saturated_ticks 위를 관절 0~6별로 (공백 구분 7개). 어느 관절이 한계인지

참고용

열 뜻
attempt 세션 누적 시도 번호. 로그 디렉토리를 이어 쓰면 episode_id와 안 맞을 수 있음
task, quantize, quantize_mode 실행 설정 확인용 (디렉토리 이름에 이미 있음)
raw_chunk_len, quantized_chunk_len 16 → 4 등. 설정대로 압축됐는지 확인
sec_per_tick duration_sec / ticks. 파생값
policy_queries, avg_inference_sec 추론 횟수·시간
timed_out 1이면 max_timesteps에 걸려 끝난 것 = 실패. duration_sec을 쓰면 안 됨

토크는 이 3열만 둡니다. chunk별 집계는 tracking.tsv의 torque_frac + chunk_idx로, 균일 dt 집계는 nominal_torque_frac으로 그대로 재계산되므로 열을 따로 두지 않습니다. 토크 계산에 쓴 URDF·payload는 실행 시작할 때 콘솔에 [torque] saturation check backend: ... 로 찍힙니다.

results/tracking.tsv — one row per tick

핵심

열 뜻
episode_id, t_step 어느 에피소드의 몇 번째 tick
chunk_idx, idx_in_chunk, chunk_len 어느 action chunk의 몇 번째 waypoint
cmd_j0..j6 명령한 관절 위치 (rad). 토크 추정의 입력
meas_j0..j6 이 tick 시작 시점의 실측 관절 위치
lag_l2 ‖이전 명령 − 지금 실측‖. 압축할수록 커지면 로봇이 끌려간 것
step_dist_l2 ‖지금 명령 − 이전 명령‖. 한 tick에 요구된 이동량 (q에 비례해 커짐)
tick_dt 이전 명령과의 실제 시간 간격 (s)
torque_frac 이 tick의 max(|τ|/정격). > 1 인 tick이 문제 지점

참고용

열 뜻
attempt 위와 같음
cmd_gripper, meas_gripper 그리퍼. 관절 토크와 무관
lag_max lag_l2의 관절별 최댓값
torque_saturated torque_frac > 1 을 0/1로. 필터링 편의용
nominal_torque_frac 균일 dt 진단용

각 에피소드의 앞 2 tick은 torque_*가 빈 칸입니다 — 가속도를 만들려면 이전 명령 두 개가 필요하고, 그때는 아직 없습니다. 0이 아니라 빈 칸인 이유는 0이 "힘이 전혀 필요 없었다"로 읽히기 때문입니다.

나머지 (전부 참고용)

파일 뜻
results/eval.tsv task별 누적 평균 한 줄. episodes.tsv에서 다시 계산할 수 있으므로 굳이 볼 필요 없음. 에피소드를 지우면 틀린 값이 남음
results/process_tracking.tsv Best-of-N 점수 기록. q1~q4는 Best-of-N을 안 쓰므로 대부분 빈 칸
results/outputs_NNN.json 정책이 뱉은 원본 출력 (chunk 모양 등). 디버깅용
videos/NNN.mp4 외부+손목 카메라 합친 영상, 15 fps
../result/<variant>.csv variant 간 비교용 요약. timestamp, task, success, steps, wall_time_sec, torque_saturated_ticks, max_torque_frac_of_rated

주의

  • 모든 파일이 append이고 헤더는 파일 생성 시 한 번만 씁니다. 열이 다른 옛 파일에 이어 쓰면 값이 밀리므로, 스키마가 바뀌면 옛 파일을 옮기고 새로 시작하세요.
  • episodes.tsv 행은 rollout이 끝나고 성공 여부를 입력한 뒤에 기록됩니다. 중간에 죽으면 영상과 tracking.tsv 행만 남아 짝이 안 맞습니다 — delete_episode.py로 정리.
  • 이미 기록된 에피소드의 토크 값은 tracking.tsv만 있으면 로봇 없이 다시 계산할 수 있습니다 (torque_backfill.py). 토크는 명령 궤적에 역동역학을 적용한 추정치이고, 로봇이 실제로 낸 토크 측정값이 아닙니다.
Downloads last month
165