The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
id: string
records: int64
preview_type: string
source_publisher: string
source_url: string
rights_evidence: list<item: string>
child 0, item: string
asset_rights_check: string
limitations: list<item: string>
child 0, item: string
diagram_type: string
source_page: string
technical_domain: string
component_or_concept_candidates: list<item: string>
child 0, item: string
rights_basis: string
asset_title: string
preview_status: string
benchmark_tasks: list<item: string>
child 0, item: string
source_record_id: string
to
{'source_record_id': Value('string'), 'source_publisher': Value('string'), 'source_url': Value('string'), 'source_page': Value('string'), 'asset_title': Value('string'), 'diagram_type': Value('string'), 'technical_domain': Value('string'), 'component_or_concept_candidates': List(Value('string')), 'benchmark_tasks': List(Value('string')), 'rights_basis': Value('string'), 'preview_status': Value('string')}
because column names don't match
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 2398, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_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
id: string
records: int64
preview_type: string
source_publisher: string
source_url: string
rights_evidence: list<item: string>
child 0, item: string
asset_rights_check: string
limitations: list<item: string>
child 0, item: string
diagram_type: string
source_page: string
technical_domain: string
component_or_concept_candidates: list<item: string>
child 0, item: string
rights_basis: string
asset_title: string
preview_status: string
benchmark_tasks: list<item: string>
child 0, item: string
source_record_id: string
to
{'source_record_id': Value('string'), 'source_publisher': Value('string'), 'source_url': Value('string'), 'source_page': Value('string'), 'asset_title': Value('string'), 'diagram_type': Value('string'), 'technical_domain': Value('string'), 'component_or_concept_candidates': List(Value('string')), 'benchmark_tasks': List(Value('string')), 'rights_basis': Value('string'), 'preview_status': Value('string')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Patent & Engineering Diagram Reasoning — Authentic Engineering Preview
NKO Data Labs · design-partner market probe · target commercial release: 25k+ figure-linked examples
A multimodal engineering-data concept linking technical figures to structured component/concept and description context for diagram understanding and technical reasoning.
Preview status
This repository now contains a small authentic source-backed engineering-diagram preview using selected U.S. Geological Survey technical diagrams. Every selected official USGS media page explicitly states Sources/Usage: Public Domain, and the preview retains the exact source page plus rights basis for independent verification. The downloadable preview is structured metadata and official source links; it does not copy image bytes into the dataset.
The broader product hypothesis still includes patent figures, but patent-specific expansion is a separate future source path. As of 2026-06-18, USPTO Open Data Portal access requires a valid USPTO.gov account/authenticated API route, and patent drawings can contain copyright/mask-work notices or embedded trademark risk that require item-level filtering. No claim is made that patent-figure coverage or the target 25k+ corpus is currently available.
Intended ML tasks
- diagram-to-description retrieval
- component/concept extraction
- diagram-grounded question answering
- technical relation extraction
- multimodal engineering evaluation
Preview fields
source record identifier, official source page, asset title, diagram type, technical domain, component/concept candidates, benchmark tasks, rights basis and preview status.
Provenance and rights
Primary preview source: official U.S. Geological Survey media pages. USGS states that USGS-authored or produced data and information are considered U.S. public domain, while also warning that some third-party material can be separately copyrighted. For this preview, NKO Data Labs selected only exact media pages whose own Sources/Usage field explicitly says Public Domain.
Examples include public-domain USGS block diagrams, cross sections, system diagrams and conceptual models covering tectonics, geophysics, hydrogeology, flood risk, water infrastructure and hydrometry.
Delivery target
JSONL/Parquet metadata plus permitted referenced assets where later packaging is justified, documented schema, QA report and task splits.
Commercial access
Indicative commercial licensing starts at USD 2,500 for non-exclusive internal use. Enterprise/model-training terms depend on scope and asset rights. Design partners can request a rights-cleared engineering-diagram pilot.
Limitations
This is a small authentic engineering-diagram market-validation preview, not the target commercial corpus. It does not currently include downloadable patent-figure image files. No claim is made that the target 25k+ commercial corpus or patent-specific coverage is currently available.
- Downloads last month
- 34