Dataset Viewer issue

#3
by theblackcat102 - opened

The dataset viewer is not working.

Error details:

Error code:   ConfigNamesError
Exception:    DatasetWithScriptNotSupportedError
Message:      The dataset viewer doesn't support this dataset because it runs arbitrary python code. Please open a discussion in the discussion tab if you think this is an error and tag 

@lhoestq

	 and 

@severo

	.
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/dataset/config_names.py", line 64, in compute_config_names_response
                  for config in sorted(get_dataset_config_names(path=dataset, token=hf_token))
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/inspect.py", line 351, in get_dataset_config_names
                  dataset_module = dataset_module_factory(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/load.py", line 1511, in dataset_module_factory
                  raise e1 from None
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/load.py", line 1480, in dataset_module_factory
                  return HubDatasetModuleFactoryWithScript(
                File "/src/services/worker/src/worker/utils.py", line 400, in raise_unsupported_dataset_with_script_or_init
                  raise DatasetWithScriptNotSupportedError(
              libcommon.exceptions.DatasetWithScriptNotSupportedError: The dataset viewer doesn't support this dataset because it runs arbitrary python code. Please open a discussion in the discussion tab if you think this is an error and tag 

@lhoestq

	 and 

@severo

	.

It seems that automated builder doesn't detect the train split for "*_dev.csv", only test and validation was detected. See the current README.md for original tested details. So far I have tried changing the postfix to "_train" and it doesn't work either.

Any recommendations on how to config the "dev" set for these many subsets?

cc @albertvillanova @lhoestq @severo .

Here's one of the subject config, the dataset automated builder doesn't seems to detect datafiles as redundant parameters and ignore train split. However, I have cross checked with other datasets and there's no syntax error or whatsoever.

configs:
  - config_name: engineering_math
    datafiles:
      - split: train
        path: data/engineering_math_dev.csv
      - split: validation
        path: data/engineering_math_val.csv
      - split: test
        path: data/engineering_math_test.csv

Due to security reasons, the viewer is no longer supported on datasets containing a Python loading script.

Note that the dataset can be loaded, although the viewer does not work. You could disable the viewer.

Alternatively, you could convert your dataset to a format which the viewer is supported on, for example, to Parquet by using this Space: https://huggingface.co/spaces/albertvillanova/convert-dataset-to-parquet

Found the root of issues : typo in data_files instead of datafiles. Everything is now working correctly

ikala-ray changed discussion status to closed

oh... ok

Sign up or log in to comment