Datasets:

cwLeeDev
/

aiflow-math-ink-0.9-dataset

Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
schema: string
sample_id: string
writer_id: string
session_group: string
quality_status: string
source_partition: string
target_display: string
target_cells: list<item: struct<formula_id: string, token: string>>
  child 0, item: struct<formula_id: string, token: string>
      child 0, formula_id: string
      child 1, token: string
target_relations: list<item: struct<type: string, from: int64, to: int64>>
  child 0, item: struct<type: string, from: int64, to: int64>
      child 0, type: string
      child 1, from: int64
      child 2, to: int64
formula_cells: list<item: null>
  child 0, item: null
ownership_status: string
label_status: string
canvas: struct<width: double, height: double>
  child 0, width: double
  child 1, height: double
strokes: list<item: struct<stroke_id: int64, order: int64, points: list<item: struct<x: double, y: double, po (... 157 chars omitted)
  child 0, item: struct<stroke_id: int64, order: int64, points: list<item: struct<x: double, y: double, pointer_type: (... 145 chars omitted)
      child 0, stroke_id: int64
      child 1, order: int64
      child 2, points: list<item: struct<x: double, y: double, pointer_type: string, tilt_x: int64, tilt_y: int64, twist: i (... 97 chars omitted)
          child 0, item: struct<x: double, y: double, pointer_type: string, tilt_x: int64, tilt_y: int64, twist: int64, tange (... 85 chars omitted)
              child 0, x: double
              child 1, y: double
              child 2, pointer_type: string
              child 3, tilt_x: int64
              child 4, tilt_y: int64
              child 5, twist: int64
              child 6, tangential_pressure: int64
              child 7, width: double
              child 8, height: double
              child 9, t_ms: double
              child 10, pressure: double
stroke_count: int64
point_count: int64
pressure_available: bool
labels: list<item: string>
  child 0, item: string
groups: list<item: list<item: int64>>
  child 0, item: list<item: int64>
      child 0, item: int64
accepted: bool
to
{'schema': Value('string'), 'sample_id': Value('string'), 'writer_id': Value('string'), 'groups': List(List(Value('int64'))), 'labels': List(Value('string')), 'accepted': Value('bool')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1816, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                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 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              schema: string
              sample_id: string
              writer_id: string
              session_group: string
              quality_status: string
              source_partition: string
              target_display: string
              target_cells: list<item: struct<formula_id: string, token: string>>
                child 0, item: struct<formula_id: string, token: string>
                    child 0, formula_id: string
                    child 1, token: string
              target_relations: list<item: struct<type: string, from: int64, to: int64>>
                child 0, item: struct<type: string, from: int64, to: int64>
                    child 0, type: string
                    child 1, from: int64
                    child 2, to: int64
              formula_cells: list<item: null>
                child 0, item: null
              ownership_status: string
              label_status: string
              canvas: struct<width: double, height: double>
                child 0, width: double
                child 1, height: double
              strokes: list<item: struct<stroke_id: int64, order: int64, points: list<item: struct<x: double, y: double, po (... 157 chars omitted)
                child 0, item: struct<stroke_id: int64, order: int64, points: list<item: struct<x: double, y: double, pointer_type: (... 145 chars omitted)
                    child 0, stroke_id: int64
                    child 1, order: int64
                    child 2, points: list<item: struct<x: double, y: double, pointer_type: string, tilt_x: int64, tilt_y: int64, twist: i (... 97 chars omitted)
                        child 0, item: struct<x: double, y: double, pointer_type: string, tilt_x: int64, tilt_y: int64, twist: int64, tange (... 85 chars omitted)
                            child 0, x: double
                            child 1, y: double
                            child 2, pointer_type: string
                            child 3, tilt_x: int64
                            child 4, tilt_y: int64
                            child 5, twist: int64
                            child 6, tangential_pressure: int64
                            child 7, width: double
                            child 8, height: double
                            child 9, t_ms: double
                            child 10, pressure: double
              stroke_count: int64
              point_count: int64
              pressure_available: bool
              labels: list<item: string>
                child 0, item: string
              groups: list<item: list<item: int64>>
                child 0, item: list<item: int64>
                    child 0, item: int64
              accepted: bool
              to
              {'schema': Value('string'), 'sample_id': Value('string'), 'writer_id': Value('string'), 'groups': List(List(Value('int64'))), 'labels': List(Value('string')), 'accepted': Value('bool')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1683, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1869, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

schema string	sample_id string	writer_id string	groups list	labels list	accepted bool
aiflow-public-ownership/v1	aiflow_0100	writer_004	[ [ 0 ], [ 1, 2 ], [ 3 ], [ 4, 5 ], [ 6 ] ]	[ "1", "+", "1", "=", "2" ]	true
aiflow-public-ownership/v1	aiflow_0101	writer_004	[ [ 0 ], [ 1, 2 ], [ 3 ] ]	[ "3", "+", "9" ]	true
aiflow-public-ownership/v1	aiflow_0102	writer_004	[ [ 0, 1 ], [ 2, 3 ], [ 4 ], [ 5, 6 ], [ 7, 8 ] ]	[ "5", "\\times", "1", "=", "5" ]	true
aiflow-public-ownership/v1	aiflow_0103	writer_004	[ [ 0 ], [ 1, 2 ], [ 3 ] ]	[ "9", "+", "3" ]	true
aiflow-public-ownership/v1	aiflow_0104	writer_004	[ [ 0 ], [ 1 ], [ 2, 3 ], [ 4, 5 ], [ 6, 7 ], [ 8 ], [ 9, 10 ] ]	[ "1", "2", "+", "5", "=", "1", "7" ]	true
aiflow-public-ownership/v1	aiflow_0105	writer_004	[ [ 0 ], [ 1 ], [ 2, 3 ], [ 4 ] ]	[ "1", "2", "+", "8" ]	true
aiflow-public-ownership/v1	aiflow_0106	writer_004	[ [ 0 ], [ 1, 2 ], [ 3, 4 ], [ 5 ], [ 6 ] ]	[ "(", "x", "+", "9", ")" ]	true
aiflow-public-ownership/v1	aiflow_0107	writer_004	[ [ 0 ], [ 1 ], [ 2 ] ]	[ "b", "/", "3" ]	true
aiflow-public-ownership/v1	aiflow_0005	writer_001	[ [ 0 ], [ 1, 2 ], [ 3 ] ]	[ "0", "+", "0" ]	true
aiflow-public-ownership/v1	aiflow_0006	writer_001	[ [ 0 ], [ 1, 2 ], [ 3 ] ]	[ "1", "+", "5" ]	true
aiflow-public-ownership/v1	aiflow_0007	writer_001	[ [ 0 ], [ 1, 2 ], [ 3 ], [ 4, 5 ], [ 6 ] ]	[ "6", "\\times", "0", "=", "0" ]	true
aiflow-public-ownership/v1	aiflow_0008	writer_001	[ [ 0, 1 ], [ 2 ], [ 3 ], [ 4, 5 ], [ 6 ] ]	[ "7", "-", "6", "=", "1" ]	true
aiflow-public-ownership/v1	aiflow_0009	writer_001	[ [ 0 ], [ 1 ], [ 2 ], [ 3 ] ]	[ "1", "0", "-", "3" ]	true
aiflow-public-ownership/v1	aiflow_0010	writer_001	[ [ 0 ], [ 1 ], [ 2, 3 ], [ 4 ] ]	[ "1", "0", "+", "8" ]	true
aiflow-public-ownership/v1	aiflow_0011	writer_001	[ [ 0 ], [ 1 ], [ 2 ], [ 3 ] ]	[ "1", "2", "-", "1" ]	true
aiflow-public-ownership/v1	aiflow_0012	writer_001	[ [ 0 ], [ 1, 2, 3 ], [ 4 ], [ 5, 6 ], [ 7 ] ]	[ "3", "\\div", "1", "=", "3" ]	true
aiflow-public-ownership/v1	aiflow_0013	writer_001	[ [ 0 ], [ 1 ], [ 2, 3, 4 ], [ 5 ], [ 6, 7 ], [ 8 ] ]	[ "3", "0", "\\div", "6", "=", "5" ]	true
aiflow-public-ownership/v1	aiflow_0014	writer_001	[ [ 0, 1 ], [ 2 ], [ 3, 4, 5 ], [ 6 ], [ 7, 8 ], [ 9 ] ]	[ "4", "8", "\\div", "8", "=", "6" ]	true
aiflow-public-ownership/v1	aiflow_0015	writer_001	[ [ 0 ], [ 1 ], [ 2, 3 ] ]	[ "y", "/", "4" ]	true
aiflow-public-ownership/v1	aiflow_0016	writer_001	[ [ 0 ], [ 1, 2 ], [ 3, 4 ] ]	[ "b", "+", "4" ]	true
aiflow-public-ownership/v1	aiflow_0017	writer_001	[ [ 0 ], [ 1 ], [ 2, 3 ] ]	[ "m", "/", "4" ]	true
aiflow-public-ownership/v1	aiflow_0018	writer_001	[ [ 0 ], [ 1 ], [ 2, 3 ], [ 4 ], [ 5 ] ]	[ "(", "n", "+", "5", ")" ]	true
aiflow-public-ownership/v1	aiflow_0019	writer_001	[ [ 0, 1 ], [ 2 ], [ 3, 4 ], [ 5 ], [ 6, 7 ], [ 8 ] ]	[ "f", "(", "y", ")", "+", "2" ]	true
aiflow-public-ownership/v1	aiflow_0020	writer_001	[ [ 0, 1 ], [ 2 ], [ 3, 4 ], [ 5 ], [ 6, 7 ], [ 8, 9 ] ]	[ "f", "(", "y", ")", "+", "4" ]	true
aiflow-public-ownership/v1	aiflow_0111	writer_001	[ [ 0 ], [ 1 ], [ 2 ], [ 3, 4 ], [ 5 ] ]	[ "2", "-", "0", "=", "2" ]	true
aiflow-public-ownership/v1	aiflow_0112	writer_001	[ [ 0 ], [ 1, 2 ], [ 3 ], [ 4, 5 ], [ 6 ] ]	[ "6", "\\times", "0", "=", "0" ]	true
aiflow-public-ownership/v1	aiflow_0113	writer_001	[ [ 0 ], [ 1 ], [ 2 ], [ 3, 4 ], [ 5 ] ]	[ "7", "-", "1", "=", "6" ]	true
aiflow-public-ownership/v1	aiflow_0114	writer_001	[ [ 0, 1 ], [ 2, 3 ], [ 4 ] ]	[ "8", "+", "9" ]	true
aiflow-public-ownership/v1	aiflow_0115	writer_001	[ [ 0 ], [ 1, 2, 3 ], [ 4 ], [ 5, 6 ], [ 7 ] ]	[ "1", "\\div", "1", "=", "1" ]	true
aiflow-public-ownership/v1	aiflow_0116	writer_001	[ [ 0, 1 ], [ 2, 3 ], [ 4 ] ]	[ "x", "+", "6" ]	true
aiflow-public-ownership/v1	aiflow_0117	writer_001	[ [ 0 ], [ 1, 2 ], [ 3 ] ]	[ "b", "+", "7" ]	true
aiflow-public-ownership/v1	aiflow_0082	writer_003	[ [ 0 ], [ 1, 2 ], [ 3 ], [ 4, 5 ], [ 6 ] ]	[ "2", "+", "0", "=", "2" ]	true
aiflow-public-ownership/v1	aiflow_0083	writer_003	[ [ 0, 1 ], [ 2 ], [ 3 ] ]	[ "5", "-", "0" ]	true
aiflow-public-ownership/v1	aiflow_0084	writer_003	[ [ 0, 1 ], [ 2, 3 ], [ 4 ] ]	[ "5", "\\times", "0" ]	true
aiflow-public-ownership/v1	aiflow_0085	writer_003	[ [ 0, 1 ], [ 2, 3 ], [ 4 ], [ 5, 6 ], [ 7, 8 ] ]	[ "5", "+", "2", "=", "7" ]	true
aiflow-public-ownership/v1	aiflow_0086	writer_003	[ [ 0, 1 ], [ 2, 3 ], [ 4, 5 ], [ 6, 7 ], [ 8 ], [ 9 ] ]	[ "7", "+", "4", "=", "1", "1" ]	true
aiflow-public-ownership/v1	aiflow_0087	writer_003	[ [ 0 ], [ 1 ], [ 2 ], [ 3, 4 ] ]	[ "1", "0", "-", "7" ]	true
aiflow-public-ownership/v1	aiflow_0088	writer_003	[ [ 0 ], [ 1 ], [ 2 ], [ 3 ], [ 4, 5 ], [ 6 ], [ 7 ] ]	[ "1", "2", "-", "0", "=", "1", "2" ]	true
aiflow-public-ownership/v1	aiflow_0089	writer_003	[ [ 0 ], [ 1 ], [ 2, 3, 4 ], [ 5, 6 ], [ 7, 8 ], [ 9 ] ]	[ "3", "0", "\\div", "5", "=", "6" ]	true
aiflow-public-ownership/v1	aiflow_0090	writer_003	[ [ 0 ], [ 1, 2 ], [ 3, 4, 5 ], [ 6, 7 ], [ 8, 9 ], [ 10, 11 ] ]	[ "2", "8", "\\div", "7", "=", "4" ]	true
aiflow-public-ownership/v1	aiflow_0091	writer_003	[ [ 0 ], [ 1, 2 ], [ 3 ] ]	[ "y", "=", "3" ]	true
aiflow-public-ownership/v1	aiflow_0092	writer_003	[ [ 0, 1 ], [ 2, 3 ], [ 4 ], [ 5, 6 ], [ 7 ] ]	[ "4", "\\times", "a", "+", "1" ]	true
aiflow-public-ownership/v1	aiflow_0093	writer_003	[ [ 0 ], [ 1, 2 ], [ 3, 4 ] ]	[ "a", "=", "5" ]	true
aiflow-public-ownership/v1	aiflow_0094	writer_003	[ [ 0 ], [ 1, 2 ], [ 3 ], [ 4, 5 ], [ 6 ] ]	[ "3", "\\times", "b", "+", "1" ]	true
aiflow-public-ownership/v1	aiflow_0095	writer_003	[ [ 0 ], [ 1 ], [ 2, 3 ], [ 4, 5 ], [ 6 ] ]	[ "(", "b", "+", "4", ")" ]	true
aiflow-public-ownership/v1	aiflow_0096	writer_003	[ [ 0, 1 ], [ 2, 3 ], [ 4 ], [ 5, 6 ], [ 7 ] ]	[ "5", "\\times", "m", "+", "1" ]	true
aiflow-public-ownership/v1	aiflow_0097	writer_003	[ [ 0 ], [ 1 ], [ 2, 3 ], [ 4 ], [ 5 ] ]	[ "(", "n", "+", "9", ")" ]	true

AIFlow Math Ink 0.9 Public Stroke Dataset

AIFlow Math Ink 0.9의 온라인 수식 필기 데이터다. 데이터 관리자는 2026-08-09까지 포함된 참여자 데이터의 공개 배포 승인을 확인했다. 원래 수집 동의 범위인 AIFlow 필기 인식 모델 학습·검증과 이번 공개 배포 승인을 함께 기록한다.

구성

파일	수식	용도
`data/formulas_valid.jsonl`	110	검수 완료 학습·분석 후보
`data/formulas_pending.jsonl`	39	시각 검수 전; 기본 학습 제외
`data/formulas_reject.jsonl`	5	검수 탈락; 기본 학습 제외
`data/ownership_train.jsonl`	47	수동 검수된 stroke-to-symbol ownership

전체 154개 수식, 11명의 dataset-local writer group, 15개 session group이다. 105건은 공개 웹 수집, 49건은 소유자 휴대폰 replay다. 문자별 수량과 파일 SHA-256은 dataset_info.json에 있다.

레코드

각 formula row는 다음을 보존한다.

원본 stroke 순서와 point 순서
x, y, formula-relative t_ms
pressure·tilt·contact geometry가 있는 경우 해당 센서값
canvas 크기, 목표 token/관계, 검수 상태
재현 가능한 dataset-local writer_id, session_group

ownership_train.jsonl은 sample_id, 문자 label, 원본 stroke index group을 제공한다.

개인정보 처리

공개본에서 contributor/session/prompt 원본 ID, 기록 시각, IP, 원본 경로, PNG data URL, epoch 기반 시간 원점과 raw pressure를 제거했다. writer/session 값은 비공개 매핑과 분리된 dataset-local 별칭이다. 각 수식의 첫 point를 t_ms=0으로 이동했다.

검증

python scripts/validate_public_dataset09.py . --expected-records 154 --expected-ownership 47

검증기는 154건·quality split·47 ownership rows·SHA-256·금지 필드 부재·stroke/point count·시간 원점·ownership 정합성을 확인한다.

라이선스와 제한

DATA_USE_NOTICE.md를 따른다. 참여자 동의 및 데이터 관리자 확인에 따라 공개 배포되지만, 제3자 상업 재사용을 포괄적으로 허가하는 표준 라이선스는 선언하지 않는다. 신원 추론·재식별·필적 기반 개인 프로파일링에 사용하지 말아야 한다.

알려진 한계

154개 수식, 11 writer로 작다.
class 분포가 불균형하다.
pending/reject는 학습에 자동 포함하면 안 된다.
47개 수식만 문자 단위 ownership이 검수됐다.
CROHME는 이 저장소에 포함되지 않으며 별도 비상업 연구 검증에만 사용한다.

Downloads last month: 53

Total file size:

3.95 MB