Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
episode_index
int64
length
int64
dataset_from_index
int64
dataset_to_index
int64
data/chunk_index
int64
data/file_index
int64
videos/observation.images.wrist_camera/chunk_index
int64
videos/observation.images.wrist_camera/file_index
int64
videos/observation.images.wrist_camera/from_timestamp
float64
videos/observation.images.wrist_camera/to_timestamp
float64
videos/observation.images.side_camera/chunk_index
int64
videos/observation.images.side_camera/file_index
int64
videos/observation.images.side_camera/from_timestamp
float64
videos/observation.images.side_camera/to_timestamp
float64
0
481
0
481
0
0
0
0
0
48.1
0
0
0
48.1
1
350
481
831
0
0
0
0
48.1
83.1
0
0
48.1
83.1
2
378
831
1,209
0
1
0
0
83.1
120.9
0
0
83.1
120.9
3
358
1,209
1,567
0
1
0
0
120.9
156.7
0
0
120.9
156.7
4
361
1,567
1,928
0
2
0
0
156.7
192.8
0
0
156.7
192.8
5
382
1,928
2,310
0
2
0
0
192.8
231
0
0
192.8
231
6
387
2,310
2,697
0
3
0
0
231
269.7
0
0
231
269.7
7
522
2,697
3,219
0
4
0
0
269.7
321.9
0
0
269.7
321.9
8
604
3,219
3,823
0
5
0
0
321.9
382.3
0
0
321.9
382.3
9
543
3,823
4,366
0
6
0
0
382.3
436.6
0
0
382.3
436.6
10
620
4,366
4,986
0
7
0
0
436.6
498.6
0
0
436.6
498.6
11
547
4,986
5,533
0
8
0
0
498.6
553.3
0
0
498.6
553.3
12
612
5,533
6,145
0
9
0
0
553.3
614.5
0
0
553.3
614.5
13
511
6,145
6,656
0
10
0
0
614.5
665.6
0
0
614.5
665.6
14
555
6,656
7,211
0
11
0
0
665.6
721.1
0
0
665.6
721.1
15
526
7,211
7,737
0
12
0
0
721.1
773.7
0
0
721.1
773.7
16
538
7,737
8,275
0
13
0
0
773.7
827.5
0
0
773.7
827.5
17
499
8,275
8,774
0
14
0
0
827.5
877.4
0
0
827.5
877.4
18
598
8,774
9,372
0
15
0
0
877.4
937.2
0
0
877.4
937.2
19
541
9,372
9,913
0
16
0
0
937.2
991.3
0
0
937.2
991.3
20
488
9,913
10,401
0
17
0
0
991.3
1,040.1
0
0
991.3
1,040.1
21
447
10,401
10,848
0
18
0
0
1,040.1
1,084.8
0
0
1,040.1
1,084.8
22
544
10,848
11,392
0
19
0
0
1,084.8
1,139.2
0
0
1,084.8
1,139.2
23
455
11,392
11,847
0
20
0
0
1,139.2
1,184.7
0
0
1,139.2
1,184.7
24
593
11,847
12,440
0
21
0
0
1,184.7
1,244
0
0
1,184.7
1,244
25
429
12,440
12,869
0
22
0
0
1,244
1,286.9
0
0
1,244
1,286.9
26
417
12,869
13,286
0
22
0
0
1,286.9
1,328.6
0
0
1,286.9
1,328.6
27
484
13,286
13,770
0
23
0
0
1,328.6
1,377
0
0
1,328.6
1,377
28
533
13,770
14,303
0
24
0
0
1,377
1,430.3
0
0
1,377
1,430.3
29
443
14,303
14,746
0
25
0
0
1,430.3
1,474.6
0
0
1,430.3
1,474.6
30
459
14,746
15,205
0
26
0
0
1,474.6
1,520.5
0
0
1,474.6
1,520.5
31
495
15,205
15,700
0
27
0
0
1,520.5
1,570
0
0
1,520.5
1,570
32
446
15,700
16,146
0
28
0
0
1,570
1,614.6
0
0
1,570
1,614.6
33
635
16,146
16,781
0
29
0
0
1,614.6
1,678.1
0
0
1,614.6
1,678.1
34
458
16,781
17,239
0
30
0
0
1,678.1
1,723.9
0
0
1,678.1
1,723.9
35
481
17,239
17,720
0
31
0
0
1,723.9
1,772
0
0
1,723.9
1,772
36
408
17,720
18,128
0
32
0
0
1,772
1,812.8
0
0
1,772
1,812.8
37
424
18,128
18,552
0
32
0
0
1,812.8
1,855.2
0
0
1,812.8
1,855.2
38
642
18,552
19,194
0
33
0
0
1,855.2
1,919.4
0
0
1,855.2
1,919.4
39
479
19,194
19,673
0
34
0
0
1,919.4
1,967.3
0
0
1,919.4
1,967.3
40
467
19,673
20,140
0
35
0
0
1,967.3
2,014
0
0
1,967.3
2,014
41
447
20,140
20,587
0
36
0
0
2,014
2,058.7
0
0
2,014
2,058.7
42
462
20,587
21,049
0
37
0
0
2,058.7
2,104.9
0
0
2,058.7
2,104.9
43
472
21,049
21,521
0
38
0
0
2,104.9
2,152.1
0
0
2,104.9
2,152.1
44
403
21,521
21,924
0
39
0
0
2,152.1
2,192.4
0
0
2,152.1
2,192.4
45
456
21,924
22,380
0
40
0
0
2,192.4
2,238
0
0
2,192.4
2,238
46
468
22,380
22,848
0
41
0
0
2,238
2,284.8
0
0
2,238
2,284.8
47
409
22,848
23,257
0
42
0
0
2,284.8
2,325.7
0
0
2,284.8
2,325.7
48
414
23,257
23,671
0
42
0
0
2,325.7
2,367.1
0
0
2,325.7
2,367.1
49
412
23,671
24,083
0
43
0
0
2,367.1
2,408.3
0
0
2,367.1
2,408.3
50
307
24,083
24,390
0
43
0
0
2,408.3
2,439
0
0
2,408.3
2,439
51
296
24,390
24,686
0
44
0
0
2,439
2,468.6
0
0
2,439
2,468.6
52
284
24,686
24,970
0
44
0
0
2,468.6
2,497
0
0
2,468.6
2,497
53
327
24,970
25,297
0
45
0
0
2,497
2,529.7
0
0
2,497
2,529.7
54
314
25,297
25,611
0
45
0
0
2,529.7
2,561.1
0
0
2,529.7
2,561.1
55
346
25,611
25,957
0
46
0
0
2,561.1
2,595.7
0
0
2,561.1
2,595.7
56
303
25,957
26,260
0
46
0
0
2,595.7
2,626
0
0
2,595.7
2,626
57
328
26,260
26,588
0
47
0
0
2,626
2,658.8
0
0
2,626
2,658.8
58
349
26,588
26,937
0
47
0
0
2,658.8
2,693.7
0
0
2,658.8
2,693.7
59
323
26,937
27,260
0
48
0
0
2,693.7
2,726
0
0
2,693.7
2,726
60
347
27,260
27,607
0
48
0
0
2,726
2,760.7
0
0
2,726
2,760.7
61
329
27,607
27,936
0
49
0
0
2,760.7
2,793.6
0
0
2,760.7
2,793.6
62
322
27,936
28,258
0
49
0
0
2,793.6
2,825.8
0
0
2,793.6
2,825.8
63
287
28,258
28,545
0
50
0
0
2,825.8
2,854.5
0
0
2,825.8
2,854.5
64
361
28,545
28,906
0
50
0
0
2,854.5
2,890.6
0
0
2,854.5
2,890.6
65
295
28,906
29,201
0
51
0
0
2,890.6
2,920.1
0
0
2,890.6
2,920.1
66
331
29,201
29,532
0
51
0
0
2,920.1
2,953.2
0
0
2,920.1
2,953.2
67
327
29,532
29,859
0
52
0
0
2,953.2
2,985.9
0
0
2,953.2
2,985.9
68
308
29,859
30,167
0
52
0
0
2,985.9
3,016.7
0
0
2,985.9
3,016.7
69
309
30,167
30,476
0
53
0
0
3,016.7
3,047.6
0
0
3,016.7
3,047.6
70
344
30,476
30,820
0
53
0
0
3,047.6
3,082
0
0
3,047.6
3,082
71
286
30,820
31,106
0
54
0
0
3,082
3,110.6
0
0
3,082
3,110.6
72
291
31,106
31,397
0
54
0
0
3,110.6
3,139.7
0
0
3,110.6
3,139.7
73
384
31,397
31,781
0
55
0
0
3,139.7
3,178.1
0
0
3,139.7
3,178.1
74
282
31,781
32,063
0
55
0
0
3,178.1
3,206.3
0
0
3,178.1
3,206.3
75
321
32,063
32,384
0
56
0
0
3,206.3
3,238.4
0
0
3,206.3
3,238.4
76
288
32,384
32,672
0
56
0
0
3,238.4
3,267.2
0
0
3,238.4
3,267.2
77
381
32,672
33,053
0
57
0
0
3,267.2
3,305.3
0
0
3,267.2
3,305.3
78
353
33,053
33,406
0
57
0
0
3,305.3
3,340.6
0
0
3,305.3
3,340.6
79
320
33,406
33,726
0
58
0
0
3,340.6
3,372.6
0
0
3,340.6
3,372.6
80
318
33,726
34,044
0
58
0
0
3,372.6
3,404.4
0
0
3,372.6
3,404.4
81
336
34,044
34,380
0
59
0
0
3,404.4
3,438
0
0
3,404.4
3,438
82
298
34,380
34,678
0
59
0
0
3,438
3,467.8
0
0
3,438
3,467.8
83
340
34,678
35,018
0
60
0
0
3,467.8
3,501.8
0
0
3,467.8
3,501.8
84
376
35,018
35,394
0
60
0
0
3,501.8
3,539.4
0
0
3,501.8
3,539.4
85
312
35,394
35,706
0
61
0
0
3,539.4
3,570.6
0
0
3,539.4
3,570.6
86
283
35,706
35,989
0
61
0
0
3,570.6
3,598.9
0
0
3,570.6
3,598.9
87
291
35,989
36,280
0
62
0
0
3,598.9
3,628
0
0
3,598.9
3,628
88
317
36,280
36,597
0
62
0
0
3,628
3,659.7
0
0
3,628
3,659.7
89
274
36,597
36,871
0
63
0
0
3,659.7
3,687.1
0
0
3,659.7
3,687.1
90
345
36,871
37,216
0
63
0
0
3,687.1
3,721.6
0
0
3,687.1
3,721.6
91
282
37,216
37,498
0
64
0
0
3,721.6
3,749.8
0
0
3,721.6
3,749.8
92
295
37,498
37,793
0
64
0
0
3,749.8
3,779.3
0
0
3,749.8
3,779.3
93
296
37,793
38,089
0
65
0
0
3,779.3
3,808.9
0
0
3,779.3
3,808.9
94
289
38,089
38,378
0
65
0
0
3,808.9
3,837.8
0
0
3,808.9
3,837.8
95
284
38,378
38,662
0
66
0
0
3,837.8
3,866.2
0
0
3,837.8
3,866.2
96
314
38,662
38,976
0
66
0
0
3,866.2
3,897.6
0
0
3,866.2
3,897.6
97
311
38,976
39,287
0
67
0
0
3,897.6
3,928.7
0
0
3,897.6
3,928.7
98
279
39,287
39,566
0
67
0
0
3,928.7
3,956.6
0
0
3,928.7
3,956.6
99
311
39,566
39,877
0
68
0
0
3,956.6
3,987.7
0
0
3,956.6
3,987.7
End of preview. Expand in Data Studio

franka_multi_task_v1_vid (TsFile)

Apache TsFile version of Beegbrain/franka_multi_task_v1_vid.

Overview

A LeRobot v3.0 robot manipulation dataset covering 4 tasks. Each frame holds the commanded action and observed observation.state joint positions; camera views are stored as videos in the original dataset.

  • Episodes: 200
  • Frames: 75,843
  • Sampling rate: 10 fps
  • Tasks: 4
  • Split: a single train split (0:200)
  • Robot: not declared in the source metadata

Schema (TsFile structure)

All episodes share one TsFile with episode_index and task_index as TAG columns; query a single episode with WHERE episode_index = N.

  • Time (INT64, milliseconds) — round(timestamp * 1000); the source timestamp column is dropped (it equals Time / 1000).
  • episode_index (TAG) — device dimension.
  • task_index (TAG) — device dimension.
  • frame_index (INT64) — measurement.
  • sample_index (INT64) — measurement.
  • observation_state_0..7 (FLOAT) — measured joint positions and gripper.
  • action_0..7 (FLOAT) — commanded joint positions and gripper.

The vector columns are flattened per joint (joint_0..joint_6, gripper per the source meta/info.json feature names).

Usage

Install the Apache TsFile Python SDK (pip install tsfile) and read a converted file:

from pathlib import Path
from tsfile import TsFileReader

path = Path("data/franka_multi_task_v1_vid.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())

Source & license

Downloads last month
53