Hugging Face's logo

Datasets:
shangshang
/
embodied-ai-dataset

Tags:
robotics
embodied-ai
vla
dataset
License:
Dataset card Data Studio Files
xet
 Community
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 because of a cast error
Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 7 new columns ({'avg_force_x', 'avg_force_z', 'avg_force_y', 'contact_type', 'object_material', 'n_measurements', 'max_force'}) and 9 missing columns ({'init_x', 'init_z', 'path_length', 'final_x', 'n_steps', 'init_y', 'object', 'final_y', 'final_z'}).

This happened while the csv dataset builder was generating data using

hf://datasets/shangshang/embodied-ai-dataset/force_tactile.csv (at revision 82ee681daaccf944d9e349bbfb72da02cfcf58d2), ['hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/end_effector_poses.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/force_tactile.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/joint_configurations.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/language_instructions.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/robot_trajectory.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/vla_combined_dataset.csv']

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1848, in _prepare_split_single
                  writer.write_table(table)
                  ~~~~~~~~~~~~~~~~~~^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 765, in write_table
                  self._write_table(pa_table, writer_batch_size=writer_batch_size)
                  ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 773, in _write_table
                  pa_table = table_cast(pa_table, self._schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              sample_id: int64
              task: string
              object_material: string
              n_measurements: int64
              avg_force_x: double
              avg_force_y: double
              avg_force_z: double
              max_force: double
              contact_type: string
              success: int64
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 1480
              to
              {'sample_id': Value('int64'), 'task': Value('string'), 'object': Value('string'), 'n_steps': Value('int64'), 'init_x': Value('float64'), 'init_y': Value('float64'), 'init_z': Value('float64'), 'final_x': Value('float64'), 'final_y': Value('float64'), 'final_z': Value('float64'), 'path_length': Value('float64'), 'success': Value('int64')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              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 1850, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
                  ...<4 lines>...
                  )
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 7 new columns ({'avg_force_x', 'avg_force_z', 'avg_force_y', 'contact_type', 'object_material', 'n_measurements', 'max_force'}) and 9 missing columns ({'init_x', 'init_z', 'path_length', 'final_x', 'n_steps', 'init_y', 'object', 'final_y', 'final_z'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/shangshang/embodied-ai-dataset/force_tactile.csv (at revision 82ee681daaccf944d9e349bbfb72da02cfcf58d2), ['hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/end_effector_poses.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/force_tactile.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/joint_configurations.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/language_instructions.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/robot_trajectory.csv', 'hf://datasets/shangshang/embodied-ai-dataset@82ee681daaccf944d9e349bbfb72da02cfcf58d2/vla_combined_dataset.csv']
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

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.

sample_id
int64
task
string
object
string
n_steps
int64
init_x
float64
init_y
float64
init_z
float64
final_x
float64
final_y
float64
final_z
float64
path_length
float64
success
int64
0
place
tool
34
0.483
0.352
0.307
0.246
0.189
0.213
0.302
0
1
insert
bottle
22
-0.271
0.357
0.444
-0.328
0.317
0.368
0.103
1
2
pick
key
24
0.003
0.259
0.71
-0.082
0.126
0.67
0.163
1
3
pour
cup
27
0.042
0.015
0.492
0.229
0.247
0.469
0.298
1
4
place
tool
23
-0.492
-0.237
0.33
-0.292
-0.116
0.416
0.25
1
5
place
bottle
28
0.37
0.044
0.533
0.458
0.176
0.423
0.193
1
6
insert
cloth
48
-0.426
-0.418
0.424
-0.194
-0.409
0.361
0.241
1
7
insert
cloth
26
0.268
-0.013
0.275
0.488
0.135
0.111
0.312
1
8
wipe
bottle
37
0.068
-0.176
0.526
-0.111
-0.099
0.412
0.226
1
9
insert
key
36
-0.394
0.006
0.374
-0.172
0.277
0.64
0.44
1
10
place
tool
44
-0.228
-0.363
0.423
-0.12
-0.364
0.352
0.13
1
11
pick
tool
35
-0.46
-0.422
0.563
-0.547
-0.309
0.817
0.292
1
12
place
cloth
39
0.011
0.484
0.653
0.08
0.406
0.503
0.182
1
13
pour
cloth
42
-0.426
0.483
0.76
-0.574
0.334
1.044
0.354
1
14
insert
tool
41
0.091
0.124
0.232
0.351
0.093
0.242
0.262
1
15
pick
cloth
31
-0.079
0.204
0.486
-0.358
0.222
0.298
0.337
1
16
pick
tool
29
-0.321
-0.018
0.486
-0.057
-0.073
0.599
0.292
1
17
wipe
key
44
-0.144
0.389
0.401
0.114
0.105
0.304
0.396
1
18
wipe
key
29
-0.131
-0.307
0.794
0.057
-0.085
1.037
0.378
1
19
pick
tool
37
-0.445
0.326
0.251
-0.42
0.511
0.024
0.294
0
20
place
tool
20
0.222
-0.169
0.61
0.425
0.007
0.666
0.274
1
21
insert
tool
21
0.278
-0.362
0.277
0.141
-0.476
0.554
0.33
1
22
pour
key
21
-0.346
0.114
0.33
-0.615
0.026
0.269
0.289
1
23
place
cup
36
-0.388
0.397
0.35
-0.12
0.183
0.132
0.407
1
24
place
key
27
0.366
-0.151
0.538
0.16
0.028
0.387
0.312
0
25
insert
bottle
28
-0.35
-0.333
0.611
-0.26
-0.056
0.401
0.358
1
26
insert
key
31
-0.062
-0.22
0.72
-0.287
-0.274
0.78
0.24
1
27
pour
key
25
-0.021
0.253
0.245
0.241
0.36
0.315
0.291
1
28
pour
cup
34
0.031
0.178
0.685
0.12
0.119
0.669
0.108
1
29
pour
key
33
-0.377
-0.129
0.388
-0.207
-0.428
0.191
0.396
1
30
insert
key
37
-0.1
-0
0.625
0.006
-0.258
0.348
0.393
1
31
pour
key
49
0.098
-0.192
0.5
0.077
-0.128
0.332
0.182
1
32
wipe
key
49
-0.305
-0.084
0.586
-0.031
-0.038
0.761
0.328
1
33
pick
key
23
0.194
-0.259
0.364
0.437
0.037
0.423
0.387
1
34
pour
cup
46
-0.327
0.494
0.307
-0.058
0.389
0.404
0.305
1
35
wipe
bottle
29
-0.174
-0.143
0.651
-0.032
-0.339
0.786
0.278
0
36
insert
cup
43
-0.051
0.443
0.671
0.017
0.428
0.42
0.26
1
37
place
cloth
45
0.294
0.034
0.504
0.284
-0.026
0.682
0.188
1
38
wipe
tool
42
0.484
0.323
0.262
0.195
0.268
0.029
0.376
1
39
place
cloth
20
-0.203
-0.31
0.381
-0.271
-0.608
0.568
0.358
1
40
place
cloth
31
-0.33
0.129
0.57
-0.133
-0.129
0.857
0.434
1
41
wipe
bottle
40
-0.47
0.197
0.516
-0.454
0.198
0.387
0.13
1
42
pour
bottle
34
0.167
0.371
0.507
0.463
0.195
0.749
0.42
0
43
insert
bottle
49
-0.221
-0.086
0.441
-0.072
-0.001
0.218
0.281
1
44
pour
bottle
42
0.246
0.013
0.758
0.393
-0.153
0.613
0.266
1
45
place
key
41
0.145
-0.448
0.776
0.163
-0.687
0.725
0.245
1
46
wipe
key
32
-0.018
0.438
0.683
0.247
0.239
0.486
0.386
0
47
insert
tool
48
0.496
0.228
0.552
0.199
0.343
0.733
0.366
1
48
pour
bottle
36
0.389
0.002
0.364
0.222
0.18
0.49
0.275
1
49
place
cup
27
0.156
0.063
0.399
-0.007
0.21
0.681
0.358
1
50
insert
cloth
21
-0.062
0.426
0.571
0.114
0.544
0.86
0.359
1
51
place
tool
44
-0.257
0.206
0.475
-0.313
0.167
0.482
0.069
1
52
insert
bottle
26
-0.055
-0.361
0.272
0.119
-0.316
0.373
0.206
1
53
place
tool
35
0.017
-0.115
0.461
0.209
-0.393
0.301
0.374
1
54
pour
key
34
-0.067
0.388
0.611
-0.005
0.172
0.604
0.225
1
55
place
key
35
-0.337
-0.222
0.765
-0.422
-0.508
0.497
0.401
1
56
pour
cup
43
0.04
0.281
0.318
-0.176
0.01
0.204
0.365
1
57
wipe
key
24
0.455
0.08
0.227
0.633
0.258
0.045
0.311
1
58
wipe
cloth
45
0.215
0.265
0.331
-0.064
-0.034
0.398
0.415
1
59
insert
key
23
-0.243
0.342
0.332
-0.224
0.206
0.605
0.306
1
60
pour
cloth
24
-0.474
0.255
0.536
-0.447
0.287
0.523
0.043
1
61
pour
tool
27
0.095
0.491
0.618
0.059
0.348
0.732
0.187
1
62
insert
key
29
-0.298
-0.251
0.254
-0.396
-0.455
0.445
0.296
1
63
insert
bottle
31
-0.349
0.268
0.757
-0.049
0.419
0.895
0.363
1
64
place
bottle
28
-0.265
0.035
0.278
-0.178
0.309
0.562
0.404
1
65
pick
bottle
33
0.468
0.455
0.767
0.402
0.341
0.807
0.138
1
66
place
cup
21
0.31
-0.358
0.761
0.236
-0.285
0.496
0.285
1
67
pick
bottle
23
-0.273
-0.385
0.354
-0.434
-0.148
0.108
0.377
1
68
insert
key
31
-0.002
0.181
0.512
0.051
0.31
0.449
0.154
1
69
place
tool
24
-0.213
-0.143
0.252
-0.471
0.028
0.106
0.342
1
70
place
tool
21
0.479
0.09
0.421
0.684
0.308
0.506
0.311
0
71
pick
cup
41
0.045
0.14
0.69
-0.09
0.147
0.804
0.178
1
72
wipe
bottle
33
-0.022
-0.37
0.797
-0.129
-0.634
0.889
0.3
1
73
place
tool
38
0.341
-0.298
0.327
0.499
-0.468
0.356
0.234
1
74
insert
cup
32
0.37
0.368
0.703
0.54
0.163
0.953
0.366
1
75
insert
cup
41
0.121
0.41
0.662
0.398
0.476
0.715
0.29
0
76
wipe
tool
39
0.034
0.291
0.438
-0.265
0.128
0.693
0.426
1
77
place
bottle
20
-0.458
-0.063
0.735
-0.532
-0.33
0.651
0.29
1
78
wipe
cloth
37
0.028
0.328
0.327
0.053
0.113
0.172
0.267
1
79
insert
cup
20
0.376
-0.173
0.671
0.475
-0.232
0.596
0.137
1
80
wipe
cup
28
-0.022
0.274
0.576
-0.305
0.383
0.31
0.403
1
81
place
key
30
-0.131
-0.263
0.674
-0.388
-0.268
0.517
0.301
1
82
pick
cloth
34
-0.387
-0.035
0.586
-0.428
-0.236
0.665
0.219
1
83
place
bottle
40
-0.396
0.329
0.48
-0.475
0.572
0.369
0.279
1
84
pour
tool
35
-0.232
-0.111
0.519
-0.026
0.168
0.568
0.35
1
85
pour
key
26
-0.065
-0.279
0.366
-0.333
-0.11
0.662
0.434
1
86
place
tool
25
-0.421
0.359
0.627
-0.273
0.585
0.811
0.327
1
87
place
cup
24
-0.187
-0.31
0.536
-0.214
-0.419
0.288
0.271
0
88
insert
cloth
38
-0.481
-0.475
0.264
-0.548
-0.226
0.257
0.258
1
89
insert
cloth
30
-0.198
0.047
0.39
-0.45
0.009
0.478
0.269
1
90
insert
bottle
28
0.283
-0.297
0.464
0.161
-0.12
0.39
0.227
0
91
insert
cup
37
0.486
-0.205
0.212
0.704
-0.37
0.497
0.395
1
92
insert
key
39
0.042
0.453
0.492
-0.04
0.552
0.2
0.32
1
93
pick
cloth
24
-0.38
-0.346
0.414
-0.593
-0.179
0.638
0.351
1
94
pour
cloth
21
0.178
-0.372
0.398
0.449
-0.153
0.635
0.422
1
95
pick
cup
24
0.235
-0.11
0.525
0.217
-0.164
0.263
0.268
0
96
insert
cup
32
-0.42
0.482
0.254
-0.39
0.236
0.54
0.379
1
97
pick
bottle
40
-0.067
-0.262
0.33
-0.134
-0.306
0.286
0.092
1
98
pick
key
43
-0.3
0.381
0.707
-0.428
0.513
0.567
0.231
1
99
pick
key
49
0.223
0.267
0.266
0.158
0.486
0.412
0.272
1
End of preview.

ACCESS REQUIREMENT - FOLLOW TO DOWNLOAD

This dataset requires following the author to access.

How to Access

  1. Follow @shangshang on HuggingFace: https://huggingface.co/shangshang
  2. Request access by commenting on the dataset page
  3. Once approved, you will receive download permissions

Usage Agreement

  • For research and educational purposes only
  • Do not redistribute without permission
  • Cite the dataset in your work:
@misc{shangshang_dataset_2026,
  title={Embodied AI and Medical Datasets},
  year={2026},
  url={https://huggingface.co/datasets/shangshang}
}

Embodied AI Dataset Collection

A comprehensive dataset for Vision-Language-Action (VLA) model training and embodied AI research.

Datasets Included

Dataset Samples Description
robot_trajectory.csv 1000 Robot arm trajectories for various manipulation tasks
joint_configurations.csv 800 6-DOF joint angle configurations
end_effector_poses.csv 600 End-effector pose sequences
force_tactile.csv 500 Force/tactile sensor feedback
language_instructions.csv 1000 Natural language task instructions
vla_combined_dataset.csv 500 Combined VLA training data

Task Categories

  • pick_place: Pick up object and place at target
  • push_pull: Push/pull objects
  • stacking: Stack objects
  • insertion: Insert objects into targets
  • turning: Turn knobs/dials

Robot Types

  • Franka Emika (7-DOF)
  • UR5e (6-DOF)
  • WidowX 250 (7-DOF)
  • xArm6 (6-DOF)

Usage Example 

import pandas as pd
from huggingface_hub import hf_hub_download

# Download
file_path = hf_hub_download(
    repo_id="shangshang/embodied-ai-dataset",
    filename="vla_combined_dataset.csv"
)

# Load
df = pd.read_csv(file_path)
print(df.head())

License

MIT License

Citation

If you use this dataset in your research, please cite:

@misc{embodied_ai_dataset_2026, title={Embodied AI VLA Dataset}, year={2026}, url={https://huggingface.co/datasets/shangshang/embodied-ai-dataset} }
Downloads last month
72
Total file size:
354 kB