Hugging Face's logo

Spaces:
appQQQ
/
Fin-EvalOps-v2
Running

App Files Community
Fin-EvalOps-v2

Commit History

  1. sync from GitHub: 5ed7463 support nullable LLM schema fields Running

    codex-sync
    6049a68

  2. sync from GitHub: b3d21a3 repair legacy positional score mappings

    codex-sync
    d8acecb

  3. sync from GitHub: 50e5d60 harden evaluations and scoped data analysis

    codex-sync
    0c2dfb3

  4. sync from GitHub: 4230296 harden backend and redesign UI

    codex-deploy
    6c7d261

  1. fix(evaluator): unwrap dimension_scores.item (llm_client.py)

    appQQQ
    Verified signature
    ba7f7c8

  2. fix(evaluator): unwrap dimension_scores.item (evaluator.py)

    appQQQ
    Verified signature
    960eb71

  3. feat(runs): batch concurrency + question + delete (runs.py)

    appQQQ
    Verified signature
    b78d971

  4. feat(runs): batch concurrency + question + delete (evaluator.py)

    appQQQ
    Verified signature
    5e562dc

  5. feat(runs): batch concurrency + question + delete (config.py)

    appQQQ
    Verified signature
    122a80e

  6. feat(runs): batch concurrency + question + delete (schemas.py)

    appQQQ
    Verified signature
    a16eac1

  7. fix(evaluator): bump max_tokens 8k→16k + raise on truncated (evaluator.py)

    appQQQ
    Verified signature
    b40e117

  8. fix(evaluator): bump max_tokens 8k→16k + raise on truncated (llm_client.py)

    appQQQ
    Verified signature
    3f2b5a9

  9. fix(evaluator): coerce list-vs-dict (retry 1, llm_client.py)

    appQQQ
    Verified signature
    bd46378

  10. fix(evaluator): coerce list-vs-dict (evaluator.py)

    appQQQ
    Verified signature
    5ed3eba

  11. docs: add agent.md (architecture overview + operational notes)

    appQQQ
    Verified signature
    d7093d4

  12. fix(evaluator): transpose parallel-array root_causes (evaluator.py)

    appQQQ
    Verified signature
    149ec2b

  13. fix(evaluator): transpose parallel-array root_causes (llm_client.py)

    appQQQ
    Verified signature
    7caca81

  14. fix(docker): drop COPY 数据测试集/ line — data uploaded via UI

    appQQQ
    Verified signature
    ff81368

  15. scan-disk: graceful no-op when 数据测试集/ not bundled

    appQQQ
    Verified signature
    bf42728

  16. remove 数据测试集/ (per user — only ship platform code)⁠

    appQQQ
    Verified signature
    7d79f77

  17. sync from GitHub bb1f476: fix(testsets): preserve `tags` field across all import paths⁠

    appQQQ
    Verified signature
    9d4f904

  1. fix(skills): route order — /file and /tree before greedy catch-all

    appQQQ
    Verified signature
    d3bc1ea

  2. feat(skills): add /tree endpoint for Skill Detail 5-tab layout

    appQQQ
    Verified signature
    adb8a8b

  3. refactor: rename expected_answer → agent_answer (evaluator.py)

    appQQQ
    Verified signature
    833e7d7

  4. refactor: rename expected_answer → agent_answer (testsets.py)

    appQQQ
    Verified signature
    9e31b0d

  5. refactor: rename expected_answer → agent_answer (schemas.py)

    appQQQ
    Verified signature
    8346b0d

  6. refactor: rename expected_answer → agent_answer (models.py)

    appQQQ
    Verified signature
    5e0c925

  7. refactor: rename expected_answer → agent_answer (db.py)

    appQQQ
    Verified signature
    1eb2028

  1. fix(evaluator): sanitize LLM judge output

    appQQQ
    Verified signature
    3cb568d

  2. update persistence.py (v2 fresh deploy)

    appQQQ
    Verified signature
    20fc803

  3. update config.py (v2 fresh deploy)

    appQQQ
    Verified signature
    60b9eec

  4. init: import all code from Fin-EvalOps (with cache-fix middleware)

    appQQQ
    Verified signature
    bcee114

  5. initial commit

    appQQQ
    Verified signature
    0426d65