Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: TypeError
Message: expected bytes, int found
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2397, in _iter_arrow
schema = self.features.arrow_schema
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1947, in arrow_schema
return pa.schema(self.type).with_metadata({"huggingface": json.dumps(hf_metadata)})
^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1936, in type
return get_nested_type(self)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1426, in get_nested_type
{key: get_nested_type(schema[key]) for key in schema}
~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1441, in get_nested_type
value_type = get_nested_type(schema.feature)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1430, in get_nested_type
{key: get_nested_type(schema[key]) for key in schema}
~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1429, in get_nested_type
return pa.struct(
~~~~~~~~~^
{key: get_nested_type(schema[key]) for key in schema}
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
) # however don't sort on struct types since the order matters
^
File "pyarrow/types.pxi", line 5340, in pyarrow.lib.struct
File "pyarrow/types.pxi", line 3836, in pyarrow.lib.field
new CField(tobytes(name), _type.sp_type, nullable, c_meta)
File "<stringsource>", line 15, in string.from_py.__pyx_convert_string_from_py_6libcpp_6string_std__in_string
TypeError: expected bytes, int foundNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Open Food Facts Canada Dataset
A Canada-focused subset of the Open Food Facts food product database, distributed as a Parquet file.
Dataset
| Property | Value |
|---|---|
| File | openfoodfacts_canada.parquet |
| Rows | 124,145 |
| Columns | 25 |
| Unique barcodes | 124,145 |
| Duplicate barcode rows | 0 |
| Canada coverage | 100% of records contain Canada in countries_tags |
| Image metadata | Available in images for all records |
| Front image URLs | 28,608 records (23.044% coverage) |
| Format | Parquet |
Scope
Every record contains Canada in countries_tags. The dataset is Canada-focused rather than strictly Canada-only: 6,583 records also contain other country or world tags.
Important columns
code: Product barcode, stored as a string.product_name: Product name.brands: Brand information when available.categoriesandcategories_tags: Product category information when available.ingredients_text: Ingredient text when available.nutriments: Nutritional data when available.nutriscore_gradeandnutriscore_score: Nutri-Score data when available.nova_group: NOVA group classification when available.countries_tags: Country tags used for the Canada-focused selection.images: Structured image metadata.completeness: Product data completeness value.last_modified_t: Product modification timestamp from the source data.front_image_url: Derived front-image URL when suitablefront_enimage metadata is available.
Data notes
- Barcode values must be treated as strings so leading zeroes are preserved. In this dataset, 113,135 barcode values begin with zero.
- Front image URLs are not available for every product. There are 95,537 records without a front image URL.
- Several product, ingredient, category, label, nutrition, quantity, and serving-size fields contain null values.
- This repository does not claim that every product has an image.
Source and reproduction
The included OFF_Canada_Data_Code.ipynb loads the openfoodfacts/product-database dataset from Hugging Face using the food split. It filters records whose countries_tags contain en:canada, selects the dataset columns, and derives front_image_url values from front_en image metadata when image ID and revision values are available.
The notebook is written for Google Colab. It installs datasets and duckdb, downloads the source Parquet file, creates the Canada-focused Parquet output, adds front-image URLs, and downloads the resulting file. The notebook writes an output named off_canada_with_images.parquet; the repository copy is named openfoodfacts_canada.parquet.
Repository files
- openfoodfacts_canada.parquet: Primary dataset.
- OFF_Canada_Data_Code.ipynb: Generation and reproduction notebook.
- metadata.json: Machine-readable dataset metadata.
License
No license is specified in this repository.
- Downloads last month
- 24