Dataset Viewer issue
#2
by
luodian
- 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 65, 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 1512, in dataset_module_factory
raise e1 from None
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/load.py", line 1481, 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
.
I've already made a parquet file and why it doesnt display in dataset viewer?
The information inside are:
Sometimes it takes some time until the viewer is updated.
Now the viewer is fixed:
albertvillanova
changed discussion status to
closed
hi thanks for your help!
But may I know why it can not display images?
I used the following script to convert images from path to parquet dict.
def convert_image_to_byte_dict(file_path):
file_path = os.path.join("/raid/bli/data/MME_Images", file_path)
with Image.open(file_path) as img:
img_byte_arr = io.BytesIO()
img.save(img_byte_arr, format=img.format)
img_bytes = img_byte_arr.getvalue()
return {"bytes": img_bytes, "path": None}
luodian
changed discussion status to
open
Hi ! Parquet doesn't have a native type for images, so you need to add Parquet schema metadata to set the Image() type:
import fsspec
import pyarrow.parquet as pq
from datasets import Features, Image
current_schema = pq.read_schema(fsspec.open("https://huggingface.co/datasets/Otter-AI/MME/resolve/main/mme.parquet").open())
features = Features.from_arrow_schema(current_schema)
features["image"] = Image()
schema_with_image_type = features.arrow_schema
print(schema_with_image_type)
# question_id: string
# category: string
# image: struct<bytes: binary, path: string>
# child 0, bytes: binary
# child 1, path: string
# question: string
# answer: string
# -- schema metadata --
# huggingface: '{"info": {"features": {"question_id": {"dtype": "string", "' + 199
# Now your should write your parquet file using `schema=schema_with_image_type`
that's nice, thank you!
luodian
changed discussion status to
closed
Or using datasets:
from datasets import Dataset, Image
ds = Dataset.from_parquet("https://huggingface.co/datasets/Otter-AI/MME/resolve/main/mme.parquet")
ds = ds.cast_column("image", Image())
ds.push_to_hub("Otter-AI/MME")
# or ds.to_parquet("path/to_local/mme.parquet")