search

google-research / language-table

Suite of human-collected datasets and a multi-task continuous control benchmark for open vocabulary visuolinguomotor learning.
Apache License 2.0
279 stars 25 forks source link

Raw collection datasets #28

Open trevorablett opened 1 year ago

trevorablett commented 1 year ago

Thanks for providing this fantastic dataset and congratulations on your really amazing work! In your paper, you mention that you provide the following raw collection datasets:

Simulation-Raw-Collect: This dataset consists of 6 teleoperators teleoperating a robot in simulation, following long horizon prompts. See representative prompts in Table VIII. 8318 episodes were collected with an average length of 36.8 ±15 seconds, yielding a total of 85.5 hours of raw data.

Real-World-Raw-Collect: This dataset consists of 11 teleoperators alternating over four robots, following long horizon prompts. See representative prompts in Table VIII. 23498 total episodes were collected with an average length of 9.9 minutes ±5.6 seconds, yielding a total of 3865 hours of raw data.

As far as I can tell, the repo only contains the relabelled datasets. Did I miss something, or are you planning to release the raw versions of the datasets? Thanks again.

Bailey-24 commented 11 months ago

how can you access the dataset? i meet my problem is 

Traceback (most recent call last):
  File "E:\Robot_learning\language-table-main\language_table\examples\dataset_example.py", line 57, in <module>
    app.run(main)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\absl\app.py", line 308, in run
    _run_main(main, args)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\absl\app.py", line 254, in _run_main
    sys.exit(main(argv))
  File "E:\Robot_learning\language-table-main\language_table\examples\dataset_example.py", line 42, in main
    builder = tfds.builder_from_directory(dataset_paths['language_table'])
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow_datasets\core\read_only_builder.py", line 138, in builder_from_directory
    return ReadOnlyBuilder(builder_dir=builder_dir)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow_datasets\core\logging\__init__.py", line 286, in decorator
    return function(*args, **kwargs)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow_datasets\core\read_only_builder.py", line 64, in __init__
    info_proto = dataset_info.read_proto_from_builder_dir(builder_dir)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow_datasets\core\dataset_info.py", line 1047, in read_proto_from_builder_dir
    return read_from_json(info_path)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow_datasets\core\dataset_info.py", line 1023, in read_from_json
    json_str = epath.Path(path).read_text()
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\etils\epath\abstract_path.py", line 145, in read_text
    return f.read()
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow\python\lib\io\file_io.py", line 114, in read
    self._preread_check()
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\tensorflow\python\lib\io\file_io.py", line 76, in _preread_check
    self._read_buf = _pywrap_file_io.BufferedInputStream(
tensorflow.python.framework.errors_impl.UnimplementedError: File system scheme 'gs' not implemented (file: 'gs://gresearch/robotics/language_table/0.0.1/dataset_info.json')

in this link , they said downgrade the tensorflow_datasets from 3.2.1 to 3.1.0, I did. but another problem 

Traceback (most recent call last):
  File "E:\Robot_learning\language-table-main\language_table\examples\dataset_example.py", line 57, in <module>
    app.run(main)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\absl\app.py", line 308, in run
    _run_main(main, args)
  File "C:\Users\zhujiahui\anaconda3\envs\lava\lib\site-packages\absl\app.py", line 254, in _run_main
    sys.exit(main(argv))
  File "E:\Robot_learning\language-table-main\language_table\examples\dataset_example.py", line 42, in main
    builder = tfds.builder_from_directory(dataset_paths['language_table'])
AttributeError: module 'tensorflow_datasets' has no attribute 'builder_from_directory'

Then, I ask gpt, they say I should set the GOOGLE_APPLICATION_CREDENTIALS environment variable to the path of the JSON file. do you set this json file?