Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot extract the features (columns) for the split 'train' of the config 'default' of the dataset.
Error code:   FeaturesError
Exception:    ArrowInvalid
Message:      Schema at index 1 was different: 
_readme: struct<이_파일: string, 검증: string, 필드_설명: struct<num: string, cleanup.apps: string>>
num: int64
instruct: string
background: bool
window_states: list<item: string>
meta: struct<확인: string, 성공하면: string, 실패하면: string>
verify: struct<app: string, window_title: string, engine: string, expect_slide_min: int64, expect_text: list<item: string>, checks: list<item: null>>
cleanup: struct<apps: list<item: string>, quit: bool, save: bool>
vs
_readme: struct<이_파일: string, 왜_엑셀은_필드가_다른가: string, 필드_설명: struct<instruct: string, background: string, window_states: string, verify.engine: string, verify.range: string, verify.expected: string, verify.expect_cells: string, verify.expect_row_hidden: string, verify.checks: string, cleanup: string>>
verify: extension<arrow.json>
num: int64
instruct: string
background: bool
window_states: list<item: string>
meta: struct<확인: string, 성공하면: string, 실패하면: string>
cleanup: struct<app: string, quit: bool, save: bool>
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/split/first_rows.py", line 249, in compute_first_rows_from_streaming_response
                  iterable_dataset = iterable_dataset._resolve_features()
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 4379, in _resolve_features
                  features = _infer_features_from_batch(self.with_format(None)._head())
                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2661, in _head
                  return next(iter(self.iter(batch_size=n)))
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2839, in iter
                  for key, pa_table in ex_iterable.iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2377, in _iter_arrow
                  yield from self.ex_iterable._iter_arrow()
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 580, in _iter_arrow
                  yield new_key, pa.Table.from_batches(chunks_buffer)
                                 ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^
                File "pyarrow/table.pxi", line 5039, in pyarrow.lib.Table.from_batches
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
                  raise convert_status(status)
              pyarrow.lib.ArrowInvalid: Schema at index 1 was different: 
              _readme: struct<이_파일: string, 검증: string, 필드_설명: struct<num: string, cleanup.apps: string>>
              num: int64
              instruct: string
              background: bool
              window_states: list<item: string>
              meta: struct<확인: string, 성공하면: string, 실패하면: string>
              verify: struct<app: string, window_title: string, engine: string, expect_slide_min: int64, expect_text: list<item: string>, checks: list<item: null>>
              cleanup: struct<apps: list<item: string>, quit: bool, save: bool>
              vs
              _readme: struct<이_파일: string, 왜_엑셀은_필드가_다른가: string, 필드_설명: struct<instruct: string, background: string, window_states: string, verify.engine: string, verify.range: string, verify.expected: string, verify.expect_cells: string, verify.expect_row_hidden: string, verify.checks: string, cleanup: string>>
              verify: extension<arrow.json>
              num: int64
              instruct: string
              background: bool
              window_states: list<item: string>
              meta: struct<확인: string, 성공하면: string, 실패하면: string>
              cleanup: struct<app: string, quit: bool, save: bool>

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.

OneAgent Computer Use — 리그레션 테스트 데이터셋

OneAgent(컴퓨터유즈)의 앱 조작 기능을 회귀 검증하는 시드 시나리오 모음입니다. OneAgent를 수정할 때마다 예전에 되던 기능이 조용히 깨졌는지 자동으로 잡아내는 데 쓰입니다. 매번 이 고정 세트를 심어 반복 검증하므로, 버전 관리(커밋·태그)로 "언제 무엇이 바뀌었는지"를 추적합니다.

구성

scenarios/            # 앱별 시나리오(각 파일이 그 앱의 테스트 묶음)
  excel.json          # Excel: 입력·함수(SUMIFS/IF)·정렬·필터·조건부서식·드롭다운·차트·피벗·시트관리 등
  word.json           # Word: 입력·서식·찾아바꾸기·저장/PDF
  powerpoint.json     # PowerPoint: 슬라이드·서식·표·차트·이미지·레이아웃/테마 +
                      #   디자인 파이프라인(주제→완성 덱, 사후 교정 polish, Designer 적용,
                      #   모던 커스텀 템플릿) — 전 슬라이드 VLM 디자인 게이트로 판정
  slack.json          # Slack: 메시지 전송
  notion.json         # Notion: 페이지 생성
  notepad.json        # 메모장: 파일 입력·저장
  chain.json          # 연계(다중 앱): Word 보고서 → PPT 발표자료
fixtures/             # 시나리오가 참조하는 고정 파일(변경 금지)
  VRF-IMAGE.png       # PPT 이미지 삽입용
  VRF-TEST.txt        # 메모장 입력용
templates/            # 디자인 시나리오가 적용하는 커스텀 템플릿(변경 금지)
  genon-modern-light.potx   # 모던 라이트 테마(미니멀 화이트·모던 팔레트·Pretendard)

시나리오 필드

각 시나리오(항목)는 아래 필드로 구성됩니다.

  • num — 앱별 고정 번호(테스트 선택·리포트 표기에 사용)
  • instruct — OneAgent에 줄 실제 지시문. {marker}는 실행마다 새로 심는 고유 표식(예: [VRF-T2-...])으로 치환
  • background — 배경(화면 비점유) 수행 여부 측정
  • window_states — 시작 상태(보임/최소화/닫힘/완전종료)별로 각각 실행
  • meta — 이 테스트가 무엇을 확인하는지 / 성공·실패가 각각 무슨 의미인지(사람이 읽는 설명)
  • verify — 정답(무엇을 어떻게 확인하나): engine(com/cdp/file/vlm), 기대 값·서식·개수 등. PPT 디자인 시나리오는 design_check_all(전 슬라이드를 이미지로 내보내 장마다 VLM 체크리스트 검사), design_votes, expect_theme_not_default 등 디자인 게이트 필드를 추가로 사용
  • 후처리 단계(검증기 실행 훅): apply_theme_file(커스텀 템플릿 적용) → auto_polish(사후 자동 교정) → auto_designer(PowerPoint Designer 제안 적용, M365 로그인+온라인 필요) 순서로 검증 전에 수행
  • cleanup — 검증 후 정리(오피스는 결과 저장 후 앱 종료)

_readme 로 시작하는 항목은 문서용이며 실행 대상이 아닙니다.

사용 (검증 하네스에서)

from huggingface_hub import snapshot_download
path = snapshot_download("<계정_또는_조직>/oneagent-regression",
                         repo_type="dataset", revision="v1")
# path/scenarios/*.json 를 로드해 실행

버전 관리

이 저장소는 git 기반이라 업로드(커밋)마다 이력이 남습니다. 릴리스 시점은 태그(v1, v2, …)로 고정해, 특정 버전을 재현·비교할 수 있습니다. 기능 추가·버그 수정 시 관련 시나리오를 함께 추가하는 것을 원칙으로 합니다.

변경 이력

  • 이번 릴리스: PowerPoint 디자인 파이프라인 시나리오 확장 — num 25(완성 덱 + polish + Designer 자동 적용), num 26(+ 모던 커스텀 템플릿 apply_theme_file). templates/genon-modern-light.potx 픽스처 신규 추가. 검증기 후처리 훅(apply_theme_file/auto_polish/auto_designer) 필드 도입.
  • 이전 릴리스: 앱 6종 + 연계(chain) 시나리오, 오피스 COM 검증(값·서식·차트·피벗), PPT 디자인 게이트(num 21~24), 시작 상태별 실행, 유니크 마커 체계.
Downloads last month
9