diff --git a/.gitignore b/.gitignore index 3cba56e..1e2deee 100644 --- a/.gitignore +++ b/.gitignore @@ -27,3 +27,7 @@ CLAUDE.md designreference.html # Local landing site — deployed to Cloudflare, not part of this repo /ely-eye-web/ + +bench/corpus/* +!bench/corpus/manifest.jsonl +.pi/ diff --git a/CHECKLIST.md b/CHECKLIST.md new file mode 100644 index 0000000..732ee7f --- /dev/null +++ b/CHECKLIST.md @@ -0,0 +1,170 @@ +# CHECKLIST.md — 执行进度真相 + +真相来源：`Gaps.md` 的 P0→P1→P2。从上到下一条一条做，禁止跳项。 +每完成一条：跑「验收」列的命令，把结果贴进 `证据:` 行，再打勾。 +主 agent 独自完成。禁止 subagent、worktree、新分支、并行。 +每条完成后用 `/herdr` 在右侧 split panel 起 pi review（`strata/qwen3.8-flash-next-abliterated-iq3_s` Sol High），最多 5 轮。 + +状态记号：`[ ]` 待办 `[~]` 进行中 `[x]` 已过审 `[D]` DEFERRED（代码就绪、未真跑，不计入 pass） + +--- + +## 阶段 0 — 先立量尺 + +- [x] **b-1** 下载真实语料并冻结 + - [x] b-1.1 `yubo2333/MMLongBench-Doc`：真实长文档 PDF + 人写问题 + `evidence_page_ids` + unanswerable 标记 + - [x] b-1.2 `xiaowu0162/longmemeval-cleaned`（清单原写 `xiaowu0162/longmemeval`，该 id 不存在→401）：会话记忆 QA，含 `knowledge-update` / `temporal-reasoning` / `multi-session` 类别与 `answer_session_ids`（字段实名，非 `evidence_session_ids`） + - [x] b-1.3 `openbmb/VisRAG-Ret-Test-{SlideVQA,InfoVQA,ChartQA,MP-DocVQA,ArxivQA,PlotQA}`：query→页面图像 qrels + - [x] b-1.4 `OS-Copilot/ScreenSpot-v2`：真实 UI 截图 + 指令 + bbox（实测 bbox 布局为 `[x,y,w,h]`：1271/1272 在界内 vs xyxy 61/1272） + - [x] b-1.5 `lmms-eval/Video-MME`：真实视频 + QA；清单原写「只取 short 档」，实取分片归档里全部 129 个视频（short 105 / medium 76 / long 34 题），因为 mp4 只覆盖 900 个视频中的 129 个，再砍 short 会饿死 n + - [~] b-1.6 3 个真实 git 仓库（pyparsing / regex / tokio）已下载冻结，但 b-3 的 code 集改用 `code-search-net/code_search_net`（2.2 万条 docstring→函数，天然 ground truth，n 大两个数量级）；仓库级设置挪到 b-5 + - 验收：`bench/corpus/manifest.jsonl` 28 条，每条含 `url`、`retrieved_at`、`sha256`、`annotation_source`、`license`、`files[]`；`python -m bench.manifest_check` 退出 0（实测 0）；manifest sha256 = `8fc1ed861c891d778853d41107630869f9eb267a669c5d1038e389de80114347` +- [x] **b-2** 构建多模态查询标注集（answerable n=2928） + - [x] b-2.1 统一 schema：`{query_id, query{modality,text}, gold[{source_id,region}], scope{holder,scenario,as_of}, answerable, answer, origin}`；用 `source_id` 而非 `gold_atom_ids`，因为 atom 是新系统的产物、source 才是第三方语料的稳定身份；region 按模态受限（page/bbox/turn/path+symbol/`{}`） + - [x] b-2.2 gold 映射：全部来自第三方标注（qrels / evidence_page_ids / answer_session_ids+has_answer / bbox / func_path+func_name），零模型判断 + - [x] b-2.3 配比（answerable）：文档 766≥90、图像 240≥60、截图 320≥40、视频 215≥30、代码 1237≥30、记忆 150≥90；unanswerable 2380/5308 = 44.8%≥15% + - [x] b-2.4 陷阱类别齐：知识更新 25 / 时间推理 25 / 多会话 25（LongMemEval 原生类别）+ 无答案 2380（4 种构造：cross-holder、held-out、native unanswerable、distractor） + - 验收：`python -m bench.validate_sets`（清单原名 dataset_check/manifest_check，实作合并为一个）报 `OK: 50789 sources, 2928 answerable queries, 2380 unanswerable` 退出 0；它同时校验 gold 存在性、gold holder==query holder、`observed_at <= as_of` 时间可达性、以及每个 set 文件的 SHA256 是否仍等于冻结的 manifest（重跑两次 SHA 相同由此保证） +- [ ] **b-3** 检索 benchmark：五路消融 + 延迟规模 + - [ ] b-3.1 指标实现：recall@k、MRR、nDCG@10，纯 Python，无模型 + - [ ] b-3.2 消融五路：BM25-only / dense-only / graph-only / no-rerank / fusion+rerank + - [ ] b-3.3 延迟规模曲线：1k / 10k / 100k atom，p50 与 p95 + - [ ] b-3.4 **BASELINE 报告**：在改动前的代码上跑，冻结提交 + - 验收：`python -m bench.run --config bench/configs/ablation.json` 出表；每格带 n 与 Wilson 95% 下界；`bench/reports/baseline_*.json` 入库 +- [ ] **b-4** 外部基线 + - [ ] b-4.1 仓库内 Mem0/Zep 风格同构对照：扁平 store + user/agent/run 作用域 + 无失效 + - [ ] b-4.2 同语料同硬件跑，出对照数字 + - [ ] b-4.3 附录：真 Mem0 跑法与参考值，标明其 `add()` 依赖 LLM 的不可比因素 + - 验收：`bench/reports/baseline_external_*.json`；对照基线与 ely-eye 同标注集同指标 +- [ ] **b-5** 反自证审计 + - [ ] b-5.1 单次尝试：禁 best-of-N 重摇 + - [ ] b-5.2 禁达标即停：跑满 n 才算分 + - [ ] b-5.3 每个分数旁必有 chance 基线行 + - [ ] b-5.4 producer ≠ verifier：判分器与被评对象不共享题目生成 + - 验收：`python -m bench.audit` 对故意注入的 best-of-N 用例报错退出 + +--- + +## 阶段 1 — P0：让它成为记忆 + +- [ ] **p0-1** 记忆模型加四维度 + - [ ] p0-1.1 `holder_id` + `origin`（self / observed / artifact / third_party）+ 写入方不可伪造 + - [ ] p0-1.2 双时间轴：`event_time` / `event_time_end` / `event_time_precision` / `event_time_source` + `valid_from` / `valid_to` + `supersedes` + - [ ] p0-1.3 `scenario` + `visibility`（holder / visible_to / purpose / retention_until） + - [ ] p0-1.4 `PRAGMA user_version` 迁移框架 + 旧库可重放 + - [ ] p0-1.5 视图 `v_atoms_current` 与 `v_atoms_as_of(:ts)` + - 验收：`pytest tests/test_memory_schema.py tests/test_migrations.py` 绿；时间旅行返回历史版本 +- [ ] **p0-2** 写入路径 + - [ ] p0-2.1 `remember` / `forget` / `supersede` 三写，Python 单一写者 + - [ ] p0-2.2 MCP 工具面 + HTTP `POST /api/memory{,/forget,/supersede}` + CLI 三面同源 + - [ ] p0-2.3 服务端盖 provenance：`ingested_at`、`holder_id`、`origin` 由服务端写，客户端不可传 + - [ ] p0-2.4 `forget` 级联：派生 atom、卡带成员、向量、元信念 + - 验收：`pytest tests/test_write_path.py tests/test_forget_cascade.py` 绿；端到端「记住→重启→召回」 +- [ ] **p0-3** 检索改 SQL-only + - [ ] p0-3.1 用 `db.search_fts()` 取代 `list_atoms()`，`LIMIT top_k*10` + - [ ] p0-3.2 分词修复：CJK 分词 + camelCase/标识符切分 + 前缀 + - [ ] p0-3.3 空命中返回显式 `no_match`，不返回裸空数组 + - [ ] p0-3.4 过滤器：`since/until/source/scenario/holder/modality/trust` + 游标分页 + - [ ] p0-3.5 RRF 取代 `0.55/0.35/0.10` 加权混合 + - [ ] p0-3.6 MCP 与 HTTP 收敛到同一排序实现 + - 验收：`MATCH "钥匙"` 命中「我们的钥匙放在玄关抽屉里」；`search atoms` 命中 `func SearchAtoms`；100k atom p50 <100ms；`pytest tests/test_retrieval_parity.py` 绿 +- [ ] **p0-4** 卡带变成可挂载作用域 + - [ ] p0-4.1 `mount` / `unmount` / `switch` / `append` 四动词 + - [ ] p0-4.2 检索按已挂载卡带过滤（作用域，不再是全局） + - [ ] p0-4.3 `kv_fingerprint`：model / tokenizer / chat template / adapter hash / rope / 量化 / 引擎版本 + - [ ] p0-4.4 fingerprint 不匹配则拒绝挂载，错误信息点名差异字段 + - 验收：`pytest tests/test_cartridge_scope.py tests/test_kv_fingerprint.py` 绿 +- [ ] **p0-5** holder registry + who_knows + - [ ] p0-5.1 `holders` 表 + registry CRUD + - [ ] p0-5.2 `who_knows` 工具：返回 holder 与置信度，非空结果 + - [ ] p0-5.3 转介：visited-set + 深度上限 + 终止态 `NO_HOLDER_KNOWS` + - [ ] p0-5.4 默认 answer-only 转介，不外泄原文 + - 验收：`pytest tests/test_transactive.py` 绿；钥匙测试场景通过 +- [ ] **p0-6** 证明套件去自证 + - [ ] p0-6.1 删 arena 答案入墨（`evals.py:1408-1409` 连接线、`evals.py:1417-1426` 白圈） + - [ ] p0-6.2 删 best-of-N 重摇与达标即停 + - [ ] p0-6.3 删自造矛盾场景（出题者≠判分者） + - [ ] p0-6.4 删缓存/证据图自触发式检查 + - [ ] p0-6.5 模型相关检查标 `DEFERRED`，不进 pass + - 验收：`python -m ely_eye.proofs --report` 逐条 PASS/DEFERRED/REMOVED + 理由 + +--- + +## 阶段 2 — P1：让它成为产品 + +- [ ] **p1-1** 图像进 dense + - [ ] p1-1.1 选 embedding ≤500MB，双语 + 图像；记录模型 SHA 与体积 + - [ ] p1-1.2 图像 atom 入索引；query 侧支持图像 + - [ ] p1-1.3 视觉 token 计数改 32×32 口径（1280×720 ≈ 900） + - [ ] p1-1.4 VLM 描述：代码路径就绪，标 `DEFERRED` + - 验收：无文字图像被图像查询召回有 recall 数字；`pytest tests/test_visual_tokens.py` 绿 +- [ ] **p1-2** 音频摄入 + - [ ] p1-2.1 `Modality` 加 audio/transcript；未知扩展名不再静默跳过 + - [ ] p1-2.2 ASR 接入，转写 atom 带词级时间戳 + - 验收：`pytest tests/test_audio_ingest.py` 绿；ASR 跑通或明确 `DEFERRED` +- [ ] **p1-3** adapter 加载 + - [ ] p1-3.1 按卡带/场景解析 6 个 adapter，不再只有 `runtime.py:317` + - [ ] p1-3.2 加载结果可观测：日志 + `GET /api/runtime/adapters` + - 验收：状态端点列每个 adapter 的 loaded/deferred + 原因；桩权重测试证明加载分支执行 +- [ ] **p1-4** 工程地基 + - [ ] p1-4.1 pytest 从零到有（后端） + - [ ] p1-4.2 go test 从零到有（MCP） + - [ ] p1-4.3 CI 在 Linux runner 上跑真测试 + - [ ] p1-4.4 `Dockerfile` 不再 COPY 缺失的 `PRD.md` + - [ ] p1-4.5 去掉 `RTX 4090` / `Ubuntu` 硬编码，改可配置 + - [ ] p1-4.6 补 `.sh` 脚本，与 `.ps1` 对等 + - [ ] p1-4.7 `LICENSE` + - [ ] p1-4.8 `backup` / `restore` / `doctor` + - 验收：CI 绿；干净 clone 上 `docker build` 成功；桩 `nvidia-smi` 证明设备名不再致命 +- [ ] **p1-5** 产品面 + - [ ] p1-5.1 `GET /api/search` + - [ ] p1-5.2 前端：搜索、浏览、forget、挂载卡带、时间显示、图像渲染 + - [ ] p1-5.3 `types.ts` 补回 `time` / `source_id` / `trust` / `relations` + - [ ] p1-5.4 12 个未用端点：接上或删除 + - 验收：E2E 测试覆盖五项；`grep` 无未接端点残留 +- [ ] **p1-6** 代码卫生 + - [ ] p1-6.1 `evals.py` 2413 行拆到 ≤500 行/文件 + - [ ] p1-6.2 删 dead code 与未调用函数 + - [ ] p1-6.3 删 v1/v2/v3 废弃端口 + - [ ] p1-6.4 注释压到 ≤1 行 + - 验收：`python -m bench.filesize_check` 报无文件 >500 行；vulture/grep 证据 + +--- + +## 阶段 3 — P2：规模化、安全、分层与共享 + +- [ ] **p2-1** 规模 + - [ ] p2-1.1 `sqlite-vec` `vec0` 增量索引 + - [ ] p2-1.2 MRL 截断 + int8/binary 粗排 + reranker 精排（reranker ≤2GB） + - [ ] p2-1.3 100k 实测；1M 外推表 + 假设 + - 验收：100k dense 腿 p50 <100ms 实测；新 atom 立即可查有测试 +- [ ] **p2-2** 安全与隐私 + - [ ] p2-2.1 FastAPI 鉴权 + Origin 校验 + - [ ] p2-2.2 写入信任分级与隔离：未佐证不得被引用 + - [ ] p2-2.3 注入消毒：外来文本里的指令进隔离区 + - [ ] p2-2.4 擦除级联（GDPR 风格 forget） + - 验收：无 token 写请求被拒；投毒用例在隔离区；级联测试绿 +- [ ] **p2-3** 短期/长期分层 + - [ ] p2-3.1 工作记忆层 + - [ ] p2-3.2 巩固台账 + watermark，可重放不重复计数 + - [ ] p2-3.3 `importance` / `recall_count` / `last_recalled_at` / `decay` / `archived_at` + - [ ] p2-3.4 冲突记录持久化 + - 验收：LongMemEval 风格的知识更新与时间推理两类有分数；被 supersede 的旧事实不当作现行 +- [ ] **p2-4** 共享记忆 + - [ ] p2-4.1 多用户/多 agent 作用域 + ACL + - [ ] p2-4.2 默认可见性：answer-only 转介 + - [ ] p2-4.3 能力卡：让外部 agent 能路由进 ely-eye + - 验收：两 holder 互不可见；越权被拒；一次真实跨 holder 会话证明能力卡可用 +- [ ] **p2-5** 文档单一真相 + - [ ] p2-5.1 `AGENTS.md` 重写并停止 gitignore + - [ ] p2-5.2 `README.md` 每条宣称指向可重跑命令 + - [ ] p2-5.3 `TRAINING_ARTIFACTS.md` 未跑分数删除 + - [ ] p2-5.4 `Gaps.md` 更新为已闭合状态 + - 验收：`grep` 无未跑模型分数残留；每条数字可重跑 + +--- + +## 终验 + +- [ ] **final-accept** Gaps.md §6「钥匙测试」7 条场景逐条跑通 + - 验收：核对表列场景、命令、实测数字、通过与否；未通过不标绿；benchmark 报告与基线、外部基线、chance 行并排 diff --git a/Gaps.md b/Gaps.md new file mode 100644 index 0000000..0abfc2e --- /dev/null +++ b/Gaps.md @@ -0,0 +1,347 @@ +# Gaps.md — ely-eye 现状与目标之间的差距 + +> 审查方式：全仓库通读（86 文件 / 17,146 行）+ 14 个并行审计 lane（6 个代码审计 + 8 个外部调研）， +> 每条结论都带 `file:line` 或一手来源。审查日期对应 HEAD `4cd29f9`。 +> 本文只描述差距，不含实现；末尾给出优先级与验收标准。 + +--- + +## 0. 结论先行 + +**现在的 ely-eye 是一个「带证明外壳的语料库检索器」，还不是一个「记忆系统」，更不是「转忆（transactive memory）系统」。** + +三句话概括： + +1. **数据模型里没有「记忆」的语义。** 一条 atom 只有 `text / source(文件路径) / time(入库时刻) / modality`。没有持有者、没有来源权威、没有事件时间、没有场景、没有短期/长期之分、没有重要度与衰减、没有生效/失效区间、没有可见性。所以「记忆的来源、时间、场景」这三个你点名的必要条件，**目前一个都不存在**（`schemas.py:57-70`、`db.py:61-77`）。 +2. **没有写入回路。** MCP 侧 9 个工具**全是只读**（`tools.go:90-135`，模块内 `Exec/INSERT/UPDATE/DELETE` 出现 0 次）；HTTP 只有 `ingest/path|upload`；CLI 没有 `remember/forget/search`。Agent 能取，不能记 —— 记忆只能靠人跑批摄取增长。 +3. **证明套件证明的是「工件存在」，不是「能力成立」。** 15 项检查里，被 README 当作卖点的几项（跨会话回忆、Visual HashHop、1.01M/100M 上下文、标准基准、卡带回放）实测的是：文件存在、自生成文件的哈希自洽、把答案画进图里的 4 选一、每个基准 **n=1**、以及对活库编译两次比摘要。详见 §3.9 与附录 A。 + +**同时，有几样东西是真的好的，别推翻**：PEFT 训练的反伪造证据链（权重前后哈希 + 每步梯度/更新 L2）、内容寻址的 ObjectStore、结构化的 atom 地址（`{source_id}:page:00001:region:0002`）、Go 侧只读 SQLite 的并发纪律、以及「一个命令产出机器可读证明」的 harness 形状。差距在于**测的东西**，不在于**有没有测**。 + +--- + +## 1. 目的对齐（我理解的目标，以及「做到」长什么样） + +你要的机制有两层，仓库目前只碰到了第一层的边： + +| 层 | 你的表述 | 技术上意味着什么 | 现状 | +|---|---|---|---| +| **卡带（cartridge）** | 像游戏卡带一样处理 memory/context，AI 随用随取 | 一个**可挂载、可切换、自带索引与兼容性声明**的上下文作用域；挂载即改变检索域，并且**省下真实算力**（前缀/KV 复用） | 卡带是一个**导出包**：写 parquet + zstd 的 JSON「稀疏索引」+ 只有节点的 sqlite；检索永远打全局库（`retrieval.py:22`），没有 mount/switch/scope | +| **转忆（transactive memory）** | 「我不需要知道钥匙在哪，我知道我老婆知道，我找得到我老婆就行」 | 记忆的主体是**目录（谁/什么持有）**而不是内容；核心原语是 `who_knows(topic) → holder`，然后**转述/引用**，并且允许「没人知道」这个终态 | 没有任何 holder 实体，没有 who-knows，没有转述，没有「没人知道」。只有对全部 atom 的扁平 BM25 | + +再往上是你说的**共享短期/长期记忆**，其必要条件是每条记忆自带 **来源 / 时间 / 场景** —— 见 §3.1。 + +**「做到」的可测定义**写在 §6，我把它写成 7 个场景测试（钥匙测试）。任何一项目前都跑不通。 + +--- + +## 2. 现状快照 + +``` +frontend (React, 8 panels, 只读为主) + └─ HTTP ── backend FastAPI (21 routes, 无鉴权, 无 /api/search) + ├─ IngestionService 文件 → atom（整文件 = 1 atom） + ├─ RetrievalService 每次查询 list_atoms() 全表 + 重建 BM25Okapi + ├─ ContextCompiler 关键词启发式选 profile/模态，问题原文当查询 + ├─ QwenRuntime 单轮 generate()，无 session/streaming/tool-call + ├─ CartridgeService materialize → parquet/json/sqlite + DNA=sha256(文件) + └─ ProofSuiteRunner 15 项检查 → proof_suite.json + └─ MCP (Go, stdio, 只读) ── 同一个 SQLite（mode=ro + query_only） +``` + +真实存在的能力：本地单文件 SQLite + FTS5、PDF 页/区块 atom、视频抽帧 + RapidOCR、git 仓库结构 atom、LoRA 训练与产物哈希、只读 MCP server、跨平台 Go 发布工程。 + +**仓库里没有任何测试文件**（Python 与 Go 都是 0），唯一 CI 是 Go 的 tag 触发交叉编译。`.ely_eye/` 运行时状态未提交，`PRD.md`（所有训练与证明的语料来源）被 `.gitignore` 排除 —— 也就是说**外人 clone 之后无法复现任何一条宣称**，`docker build` 甚至直接失败（`Dockerfile:16 COPY README.md PRD.md ./`）。 + +--- + +## 3. 差距清单 + +严重度：**P0** = 不解决就称不上记忆系统；**P1** = 决定产品能否落地；**P2** = 规模化/护城河。 + +### 3.1 记忆模型：没有「谁、何时、何事、对谁可见」 — **P0** + +| 缺什么 | 证据 | 后果 | +|---|---|---| +| 持有者 / 用户 / agent / 会话 | `schemas.py:57-70`、`db.py:61-77` 无 holder/owner/agent/user/scope 列 | 无法回答「谁知道」；无法做 per-agent 隔离；共享记忆无从谈起 | +| 事件时间 vs 入库时间 | `schemas.py:62` 只有 `time: datetime = Field(default_factory=utc_now)`，且**每个摄取点都省略它**（`ingestion.py:181,207,245,274`）；只读 `path.stat().st_size`，从不读 mtime / PDF `/CreationDate` / EXIF / 帧时间 | 「上周三我说过…」这类问题结构上不可答；`db.py:197 ORDER BY time DESC` 排的是入库顺序 | +| 生效/失效与取代链 | `db.py:148/158` `ON CONFLICT(atom_id) DO UPDATE SET time=excluded.time, text=excluded.text` + `db.py:173 DELETE FROM atom_fts` | 重摄即**覆盖**，旧版本不可恢复 → 没有「现在」与「过去」之分，知识更新无法测 | +| 场景 / 任务 / 项目标签 | 无字段，无检索过滤 | 你说的「场景」维度完全缺失；上下文切换无从实现 | +| 记忆类型（短期/长期/情景/语义/程序性） | 无字段 | 分层、巩固、遗忘都无处挂载 | +| 重要度 / 召回次数 / 衰减 / TTL | 无字段 | 无法排序，无法淘汰，库只增不减 | +| 来源权威与「第一手 vs 二手」 | `TrustScores{parser,ocr,model}`（`schemas.py:51-54`）测的是**解析保真度**，不是来源可信度；`ingestion.py:176-190` 给每条文本 atom 硬写 `trust=(1.0,0.0,0.0)`，无论是模型自述还是一份签署的合同；`retrieval.py`/`compiler.py` 里 grep `trust` **零命中**（存了从不参与排序） | 「我老婆知道」这条链上的**二手记忆**没有表示；无法区分「我看到的」「她告诉我的」「模型推的」 | +| 派生血缘 | `evidence_edges` 有表但**无时间戳、无 FK**（`db.py:330`），且**只写不读** | 无法回答「这条结论是从哪几条记忆推出来的」，也无法做级联删除 | +| 卡带成员关系 | `db.py:86-95` cartridges 表**没有与 atoms 的连接**；成员只存在于复制出去的 parquet 里 | 「挂载这个卡带 = 检索这个作用域」在数据层不成立 | + +**建议的最小 schema 增量**（细节与出处见附录 C）：`holder_id / origin_type / authority / event_time(+end,+precision,+source) / ingested_at / valid_from / valid_to / supersedes / scenario / memory_type / importance / last_recalled_at / visible_to / derived_from`，并把 atoms 主键改成 `(atom_id, revision)` + `ux_atoms_current` 部分唯一索引 + `v_atoms_current` / `v_atoms_as_of(:ts)` 两个视图。 + +### 3.2 写入路径：记忆只能靠人跑批 — **P0** + +- **没有 agent 可用的写原语**：MCP 9 个工具全读（`tools.go:91-131`）；`server.go:34` 的 Instructions 明确写着「do not mutate the production database」；HTTP 只有 `POST /api/ingest/path|upload`（`api.py:111,121`）；CLI 无 `remember/forget`。对照：官方 MCP memory server 9 工具里 6 个是写；Mem0 是 `add/search/get_all/update/delete/delete_all`；Letta 让 agent 自己编辑 memory block。 +- **没有写时抽取**：整文本/代码文件 = **一个** atom（`ingestion.py` 的 `{source_id}:text`），无分块、无摘要层级、无原子笔记。Mem0 是「抽取 + ADD/UPDATE/DELETE/NOOP 两阶段」，A-MEM 是原子笔记 + 链接。 +- **没有去重/合并判定**：`ObjectStore` 内容寻址（`storage.py:41-77`）已经能做字节级幂等，但关系层没利用；没有语义级 `ADD|MERGE|SKIP` 判定。 +- **没有巩固与遗忘**：无 consolidation job、无 decay、无归档、无容量预算。库单调增长。 +- **没有「谁决定什么值得记」的策略**：agent 显式写 vs 后台抽取，冲突时谁赢 —— 未定义。 + +### 3.3 读取路径：O(语料库) 的检索 + 两套互相矛盾的排序 — **P0** + +- `retrieval.py:22` `atoms = self.db.list_atoms()` 全表进 Python，`retrieval.py:53-56` 每次查询重建 `BM25Okapi` → 延迟与内存是 **O(corpus)** 而不是 O(result)。按自己的摄取速率（`video_sample_seconds=5.0` → 17,280 帧/天 → 约 6.3M atom/年）这条路必然撞墙。 +- 讽刺的是 **FTS5 已经建好且被 Go 用着**（`db.py:217-227` 的 `search_fts()` 在 Python 侧**零调用者**；`atoms.go:57` 在用）。 +- `final_score = sparse*0.55 + dense*0.35 + graph*0.10`（`retrieval.py:41`）里：**dense 默认关**（`config.py:22 dense_embeddings_enabled=False`），**graph 恒为 0**（`relations` 从未被摄取写入）→ 实际 45% 的权重贡献 0，且 UI 不显示。 +- **中文与标识符检索结构性失效**（本次已用真 schema 实测）：`db.py:79-84` 的 `atom_fts` 用 `tokenize='unicode61'` → 一整串连续中文是**一个** token。实测：文档 `我们的钥匙放在玄关抽屉里` 时，`MATCH "钥匙"` / `"玄关"` / `"抽屉"` / 甚至子串 `"钥匙放在"` **均 0 命中**；`MATCH "search"` / `"atoms"` 也 0 命中，只有 `"SearchAtoms"` 命中（camelCase 不切分）。而 `atoms.go:35-88` 空命中时**返回空切片且不报错** → agent 得出「记忆里没有」。 +- **Python 那条路同样没有中文分词**：`retrieval.py:85-86` `tokenize()` = `re.findall(r"[\w\u4e00-\u9fff]+", text)` → 同样把整串中文当一个 token（实测 query `钥匙` 与上述 doc **无交集**）→ BM25 腿 = 0。唯一救回它的是 `lexical_score`（`retrieval.py:89-95`）里的 `phrase_hits` 子串项（`token in atom.text.lower()`），给 `0.25/|q|` 的弱补偿 → 中/英标识符查询在 Python 侧是**排序极脆**，在 Go/FTS 侧是**完全查不到**。两边都缺 CJK 分词（jieba/ICU）与标识符切分（subword/camelCase/前缀），而 RapidOCR 默认中文、你也用中文说话 —— 这是最尴尬的一个 P0。 +- **没有过滤维度**：`SearchAtomsInput={query,modality,limit}`（`tools.go:24-28`）—— 没有 since/until、source、scenario、holder、trust floor、无分页游标。 +- **没有 rerank、没有查询改写/分解、没有多跳**；`retrieval.py` 无 reranker，`compiler.py` 的 `retrieval_query` 就是问题原文（planner 模型只当标签用）。 +- **两套排序**：Go 用 FTS5 bm25，Python 用 0.55/0.35/0.10 混合 → 同一个问题经不同传输层返回不同证据，不可测也不可演示。 + +### 3.4 卡带语义：是导出包，不是卡带 — **P0/P1** + +| 宣称 | 实测 | +|---|---| +| 卡带可当上下文单元使用 | 无 mount/switch/scope；`api.py:154-156,198-211` 把卡带绑到**整个语料**；`retrieval.py:22` 永远查全局库 | +| 「稀疏索引」 | `sparse_index.bm25.zst` 是 **zstd 压缩的 JSON 原文**，不是 BM25 索引（ postings/IDF/doclen 都没有） | +| 「temporal graph」 | `cartridge.py` 从 `atom.relations` 造边，而 relations 恒空 → **0 条边**；且没有 time 列 → 不 temporal | +| 「KV snapshots」 | `cartridge_assets.py:139-166` 写的是 JSON 块 `{atom_id, source, token_start, text_sha256}`，magic `ELYKIVI1`，dtype 自报 `zstd-json` → **0 个 KV 字节** | +| 「100M token 记忆胶囊」 | 4,096 个合成段，`cartridge_assets.py:198` `atom = atoms[index % len(atoms)]`（在 `for index in range(MEMORY_CAPSULE_SEGMENTS)` 里）—— 用取模把现有 atom 循环填充出 100M 的**标签** | +| 「DNA 可验证」 | `cartridge.py:83-93` 只 hash 文件字节 → 证明「字节没变」，不证明「模型/分词器/adapter/RoPE/量化一致」 | +| 「卡带回放」 | `evals.py:893-956` 的 replay = 对**活库**编译同一问题两次比摘要；卡带自己的 parquet 从未被加载 | + +**没有兼容性头**：`CartridgeManifest`（`schemas.py:176-186`）无 model hash / tokenizer hash / chat-template hash / adapter DNA / kv dtype / block size / rope 配置 / engine 版本 / 格式 semver / 签名。KV 只在产生它的 RoPE 与量化配置下有效（`profiles.py` 里 YaRN factor 4.0、`--cache-type-k q4_0` 都会改变 KV），所以**没有指纹的 KV 卡带不是没用，是不安全** —— 挂错状态下会静默产生错误的注意力而不是报错。 + +**省算力的账**（附录 B）：挂载卡带若走 re-prefill，262K ≈ 116–232s TTFT、1.01M ≈ 17.4–34.8 分钟；而真正可取的是 KV 重载：8.0 GiB / 30.8 GiB 在 PCIe/NVMe 上是 0.34–4.73s，**便宜 800–1600 倍**。今天卡带拿到的是 **0%**。 + +### 3.5 转忆与共享：没有 holder，就没有「找老婆」 — **P0** + +- **没有 holder 注册表**，因此没有 `who_knows`、没有路由、没有转述、没有「没人知道」这个终态。三层 API（MCP/HTTP/CLI）都没有这个动词。 +- **没有元记忆层**：「我相信 X 知道 Y」本身是一条有置信度、会过期的记忆。需要 `MetaBelief(holder_id, topic, basis∈{self_report, observed_success, inferred_from_artifact, third_party}, confidence, last_confirmed_at, ttl, n_confirmed, n_failed)`。 +- **没有 origin 类型**：`trust.model` 是一个连续值而不是来源类型 → 模型自己生成的内容被重新摄取后，与「我亲眼看到的」无法区分（认知科学里的 source-monitoring error；训练侧对应 model collapse）。 +- **没有转述回路的防环设计**：需要 visited-set + 深度上限 + 终态 `NO_HOLDER_KNOWS`，否则 A 问 B、B 问 A。 +- **没有 ACL / 可见性 / 冲突记录**：两个 holder 说法不一致时，今天会被检索分数**静默平均**掉，而不是把分歧呈现出来。 +- **污染已被量化**：Sleeper Memory Poisoning（arXiv 2605.15338）写入成功率 99.8%/95%，成功检索里 60–89% 被攻击者左右动作；InjecMEM（arXiv 2608.23471）同类。ely-eye 目前**摄取即信任**，无 quarantine。 +- **隐私/删除**：GDPR Art.5/16/17/25 与 CJEU C-136/21 意味着「删除」必须是**级联**：atom + 派生 atom + 卡带 + 向量 + 元记忆 + 缓存。今天连软删除都没有。 +- **反向能力卡缺失**：别人要能把问题路由**到** ely-eye，需要一张 capability/authority card；现在没有对外声明「我知道什么」。 + +### 3.6 短期/长期分层：没有工作记忆，也没有巩固 — **P0/P1** + +- `runtime.py` 是**单轮** `generate(question, context)`：无 session、无 history、无 streaming、无模型侧 tool-calling、无 agent loop。所以「短期记忆」连容器都没有。 +- 没有 短期→长期 的**晋升/降级**、没有 TTL、没有离线巩固（Letta dreaming / Generative Agents reflection 那一类）。 +- `cache_fabric.py:19` 号称多层缓存（KV snapshot / prefix / patch / NVMe），实际只有一层，且 key = `sha256(f"compiler-v2\n{profile}\n{question}")` —— **精确字符串匹配**，换个说法就 miss，命中率≈0。 + +### 3.7 多模态：无文字的图像在结构上不可见 — **P1** + +- **图像只有 OCR 文本**（`ingestion.py:202-218`，`text=ocr.text`）→ 照片/图表/白板/UI 状态的 atom 文本为空 → BM25 与 FTS5 都是 0 分，dense 默认关 → **完全检索不到**。仓库里从未产生过任何视觉 embedding；`embeddings.py:64` 查询侧只喂 `{"text": query}`，而 `_atom_payload`（`:79`）其实已经会构造 `{"image":…}` 却从不用于查询。 +- **音频不是支持的模态**：`ingestion.py:65-66` 只列 image/video 扩展，`Modality` 枚举无 audio/transcript，grep `whisper|asr|audio` 零命中。40 分钟会议今天 = 480 帧的 OCR，一句话都没转。 +- **视觉 token 记账约错 50 倍**：`ingestion.py:216` 用 `(w*h)//(224*224)`，而 Qwen3-VL 是 **32× 空间压缩 → 每 32×32=1024 像素 1 token**，并按 640 tokens/帧封顶。1280×720 帧真实约 900 token，被记成 18 → 编译出的上下文被静默超载（1 小时 720p 视频按 5s 采样 ≈ 648K 真实视觉 token，不是 13K）。 +- **视频是固定 5s 网格 + 仅 OCR**：无场景/镜头切分、无自适应 fps、无 VLM 稠密描述、无音轨、不存 `fps/frames_indices/total_num_frames/duration` → 预抽的帧无法在服务侧重新对齐时间。 +- 选 Qwen3-VL-Embedding 是对的（MMEB-V2 77.8，文/图/截图/视频统一空间，MRL 可截断），**错的是 8B 的体量对 16GB 目标**；2B 版（2.13B 参数，2048 维可截到 1024，约 2KB fp16/atom）是同一档能力的正确尺寸。 + +### 3.8 运行时与 adapter：6 个 adapter 里只有 1 个真的被加载 — **P1** + +- `runtime.py:317` 是全仓库**唯一**一处 `PeftModel.from_pretrained`，且只加载 `hme_core_lora` → HME-Vision / TTT / MTP / Router / Retrieval **从未在推理时生效**。 +- **没有 adapter-off 的对照运行**：`benchmarks.py:26` 的「Qwen3.5-9B baseline ±2%」是散文目标，从未计算 → adapter 的效果**没有 A/B**。 +- `HME-Retrieval` 训在 `Qwen3-VL-Embedding-8B` 上，但 `embeddings.py:71` 用的是**基座** `SentenceTransformer`，且 dense 默认关 → 这个 adapter 对检索的影响为 0。 +- `HME-Router` 的 `planner_model_id` 只作为标签出现；路由实际是 `compiler.py:11-17` 的硬编码预算表 + `:101-113` 的中英文关键词规则。 +- 无 streaming、无 tool-call、无多轮、`_context_images` 硬顶 6 张图且无视觉 token 预算。 + +### 3.9 验证与评测：证明外壳比能力跑得远 — **P0** + +细节见附录 A。核心四条： + +1. **答案泄漏**：Visual HashHop 的竞技场 PNG 里**把答案画出来了**（已逐行核对）—— `evals.py:1408-1409` 从源块中心向**目标块中心**画青色连线（`fill=(0,242,255), width=18` + 白色内芯），`evals.py:1417-1426` 再在目标中心画 `draw.ellipse(..., outline=(255,255,255), width=8)` 白圈；两跳版（`:1476-1487`）给中间块画青环、目标画双环。竞技场只有 1 源 + 4 候选（`render_visual_hash_arena`，920×560，tile=160），tile 由 `sha256(id)` 播种、id 只存在于 PNG 文件名里、从不参与比对。于是任务变成「哪个方格里有个白圈」（4 选一，chance 0.25）。被模仿的 Magic HashHop 是在真实 haystack 里注入随机 16 位 hash 对。 +2. **best-of-N 重掷 + 可选停止**：`evals.py:212-226` 每次尝试**重新生成一个新竞技场**，`if proof.passed: return proof`；失败的样本被丢进随机 proof_id 目录且从不拷贝（`:993-1000`）；`collect_model_visual_proofs` 一到 `pass_rate >= target` 就 `break`。分母是自适应选的 → 报告的比率无定义方差。纯随机下 3/3 全过的概率约 44%。 +3. **n=1 的基准**：`benchmarks.py:239-280` `row = next(iter(stream))` → 每个基准**只取 HF 流的第一行**；评分 `exact_or_contains_score`（`:377`）是「gold 是预测的子串就算 1.0」→ 答案 "A" 会被 "I think the answer is option J" 判对；`BenchmarkPredictionStatus`（`schemas.py:376-396`）**只有 score，没有 n**。注册了 7 个基准，只有 MMMU-Pro 与 OCRBench 真跑模型。 +4. **生产者=验证者**：`evals.py:793` 先 `CartridgeAssetService.finalize()` 再造资产，然后认证自己刚造的东西；`_cache_trace_check`（`:1258`）被同一次 `run()` 里更早的 `compile()` 调用满足；evidence-graph 门（`:2288`）被 `_runtime_generation_check` 内部的 `record_answer` 满足（`:731-736`）；memory DNA 是自哈希。 + +**并且：全仓库没有 recall@k / MRR / nDCG / hit-rate，没有任何 baseline（BM25-only、dense-only、graph-only、no-rerank、full-context-no-retrieval），没有 held-out 集，没有置信区间，没有弃权/不可答测试**（MMLongBench-Doc 本身有 22.8% 不可答题，正是这个项目需要的幻觉探针，却没跑）。训练语料是**被 gitignore 的 PRD.md**，且 `build_prd_dataset`（`training.py:63-125`）把**同一个 body 同时放进 prompt 和 answer**；router 8 条、visual 5 条是硬编码元组；`validate_dataset`（`:49-61`）只校验 schema，无 split。 + +### 3.10 工程与可复现性 — **P1** + +- **0 个测试文件**（Python + Go），9,637 行后端无测试；唯一 CI 是 Go 交叉编译。`scripts/verify.ps1` = `compileall` + `ely-eye status`（atom_count=0 也退出 0）+ `npm build`。 +- **`PRD.md` 被 gitignore**（`.gitignore:23/24`）但 `Dockerfile:16` 要 COPY 它 → 任何人的 `docker build` 都失败；同时它是全部训练数据的来源 → 宣称不可复现。 +- **硬件写死**：`evals.py:1672,1683,1780-1784` 断言 `"RTX 4090" in device_name` → 在 4080/5090/A6000 上证明直接失效（而 README 说目标是 4080 级 16GB）；`:1793-1798` 要求 unsloth/triton/xformers 版本串，而 unsloth 从未被 import。 +- **平台锁死**：`profiles.py` 一路 shell 到 `wsl.exe -d Ubuntu`（发行版名写死），16 个脚本里 15 个是 `.ps1`；`_runtime_profile_check` 因此只在 Windows+WSL+CUDA 下可能通过。 +- **存储生命周期缺失**：无 `PRAGMA user_version`、无迁移、无 rollback、无 backup/restore（`VACUUM INTO`）、无 doctor、无 forget、无磁盘预算与淘汰策略。 +- 无 LICENSE；约 35GB 权重无离线模式说明。 + +### 3.11 产品/API 面：能取不能记，能看不能操作 — **P0/P1** + +- **HTTP 没有 `/api/search`**：`retrieval.py:20` 的混合检索只能经 `POST /api/chat` 到达（`api.py:160`）→ 脚本/agent 无法「只回忆不生成」。 +- **CLI 没有 `search/ask/remember/forget`**。 +- **没有 scope key**（user/agent/run/session）、**没有 mount 动词**、没有 tool annotations（readOnlyHint/destructiveHint/idempotentHint）、search 上限 100 且无游标、无 `outputSchema`/`structuredContent`、无并发写序列化与逐项失败报告。 +- **FastAPI 无鉴权**（只有 CORS 指 localhost:5173）→ 本机任意进程都能写记忆、读全部记忆。 +- **前端是演示仪表盘而非操作台**：只有 5 个 mutation（`App.tsx:115-185`：ingestPath/uploadFiles/askQuestion/runHashHop/runProofSuite）；`api.ts` 82 行 9 个函数，无 search/browse/update/forget/delete/session/mount；**12 个后端端点前端没用**；`types.ts:111-119` 的 EvidenceAtom **丢掉了 `time / source_id / trust / relations / metadata`**（后端 `schemas.py:57-70` 都返回了）→ 溯源在 UI 里被主动扔掉；`LayoutBox.page/frame_second` 有类型但从不渲染；**`App.tsx` 里一个 `` 都没有**；8 个面板里 6 个是只读渲染器；per-leg 分数有类型（`types.ts:121-127`）从不渲染 → 那 45% 的空权重被藏起来了。 + +### 3.12 安全与隐私 — **P1/P2** + +无鉴权（上）；检索内容未做 injection 隔离（`RESEARCH.md` 自己引了 tool poisoning 2503.23278，代码里没有 quarantine）；无静态加密决策（SQLCipher vs 全盘）与备份密钥棘轮；无摄取期 PII/secret 脱敏与窗口/应用 denylist；无 retention TTL；删除无级联（§3.5）。对一个「把你见过的一切都存下来」的库，这些是上线条件而非加分项。 + +--- + +## 4. 差距的四个根因 + +1. **「命名即实现」**：论文名/机制名（YaRN、V-NSA、DVKV、KIVI、MTP、TTT、HashHop、semantic cache、temporal graph、memory capsule）先作为**字符串与文件名**进入仓库，再被证明套件认证。例：`dvkv.select_dvkv_tokens` 是纯函数，**runtime 从未调用**；`research_modules.py` 的向量是 `SHA256` 派生的假 `stable_vector`；`compiler.py` 的压缩策略字符串只是文本。 +2. **「生产者即验证者」**：证明的输入由同一模块生成（资产、胶囊、矛盾场景 `build_contradiction_scenarios()`、evidence 边、cache 条目），于是检查只能自洽，不能证伪。 +3. **「语料库而非记忆」**：抽象单位是「文件/页/帧」，不是「一次观察（observation）」+「一个持有者」+「一段有效期」。缺了这三样，来源/时间/场景就没有地方放，巩固与遗忘也没有对象。 +4. **「单用户、单轮、只读」的隐含假设**：没有 user/agent/session 列，没有多轮，没有写路径 —— 于是「共享」与「随用随取」都还无法表达。 + +--- + +## 5. 优先级与顺序 + +### P0 — 先让它成为记忆（6 项，建议按序） + +| # | 动作 | 可证伪的验收 | +|---|---|---| +| P0-1 | **记忆模型 v2**：§3.1 字段 + `(atom_id, revision)` 主键 + `v_atoms_current`/`v_atoms_as_of` + `PRAGMA user_version` 迁移 + 迁移测试 | 「上周三说的」和「现在」能给出不同答案；重摄不丢历史 | +| P0-2 | **写三件套** `remember / forget / supersede`：MCP 工具 + `POST /api/memory{,/forget,/supersede}` + CLI；写全部经 Python 单写者（Go 保持 `mode=ro`）；服务端盖 provenance；destructive 操作需确认 | agent 在一次会话内写入并在**下一会话**召回 | +| P0-3 | **检索改 SQL-only**：用 `db.search_fts()`（`LIMIT top_k*10`）取代 `list_atoms()`；修 FTS 分词（中文 CJK 分词 + camelCase/标识符切分 + 前缀）；空命中返回**显式 `no_match` 而非空数组**；加 since/until/source/scenario/holder/modality/trust 过滤 + 游标；RRF 取代加权混合 | 100k atom 下 p50 < 100ms；`MATCH "钥匙"` 能命中 `我们的钥匙放在玄关抽屉里`；搜 "search atoms" 能命中 `func SearchAtoms`；MCP 与 HTTP 同一排序 | +| P0-4 | **分块 + 写时抽取**：文本/代码分块（带 overlap 与结构边界）+ 写入时 `ADD\|MERGE\|SKIP` 语义去重 + 派生血缘入 `evidence_edges`（补时间戳与 FK） | 同一文档改写后重摄不产生重复记忆；结论可点开血缘 | +| P0-5 | **评测先立起来**：≥150 条人工标注查询的 held-out 集（附构造协议与 manifest hash）；报 recall@{1,5,10,50}/MRR/nDCG@10；5 个 baseline 全跑；固定 n、一次尝试、Wilson 95% 区间、chance 行并列；加弃权/不可答与知识更新两类任务 | 每个数字旁边有 n 与区间；删掉任一组件分数会掉 | +| P0-6 | **会话/工作记忆层**：`sessions` + `turns` 表、多轮上下文、`/api/session/*`、短期→长期晋升 job（先做「会话结束即抽取」） | 场景测试 1、2 通过（§6） | + +### P1 — 让它成为产品 + +- **P1-7 卡带真正可挂载**：`cartridge_atoms` 连接表 + `mount/unmount` 动词 + 检索作用域 + 兼容头（model/tokenizer/chat-template/adapter DNA/rope/kv dtype/engine + 格式 semver）+ ed25519 签名 + 真实稀疏索引（FTS5 或 postings）+ 增量向量索引（sqlite-vec `vec0`，MRL 截断 + int8/binary + rerank）。 +- **P1-8 转忆路由**：`holders` 注册表 + `who_knows` 工具 + 带 visited-set/深度上限/`NO_HOLDER_KNOWS` 终态的转述 + `MetaBelief` 表 + 权威/冲突记录 + 对外 capability card。 +- **P1-9 多模态补齐**：视觉 embedding（Qwen3-VL-Embedding-**2B**，用已有 `image_ref` 回溯跑一遍，无需重摄）+ ASR（faster-whisper large-v3 int8 或 parakeet-tdt-0.6b-v3，写 `Modality.transcript` 带段/词时间戳）+ 视觉 token 公式改 `pixels//1024` 并按 640/帧封顶 + 每 chunk 一次 VLM 稠密描述 + 存 `fps/frames_indices/duration`。 +- **P1-10 来源与信任**：`origin_type` 一等公民、写信任分级（candidate/quarantine → 引证前需旁证）、检索文本注入隔离、级联删除。 +- **P1-11 工程债**：pytest + Go test + 真 CI（含 100k atom 的 ingest/query 规模回归）、PRD 入库或从 Dockerfile 移除、去掉 `"RTX 4090"` 硬编码、`backup/restore/doctor`、FastAPI token 鉴权 + Origin 校验、LICENSE。 + +### P2 — 规模化与护城河 + +巩固与遗忘（consolidation ledger + watermark、decay/importance、归档）、多设备同步（CRDT/CR-SQLite）与静态加密、**真 KV 卡带**（safetensors 分片或 CacheBlend 式 chunk KV + 选择性重算）+ 把 cache 命中写进证明（`usage.cache_read_input_tokens` / vLLM prefix-cache hit）、agent loop / tool-calling / streaming、late-interaction rerank（ColPali/ColQwen）。 + +--- + +## 6. 「做到」的验收标准（钥匙测试） + +以下 7 个场景必须**全部可跑**，否则不叫记忆系统： + +1. **随手记**：我说「钥匙放玄关抽屉了」→ agent 调 `remember` → 下次问「我钥匙在哪」→ 命中该记忆，答案带 **事件时间 / 来源=我说的 / 场景=家** 并引用 atom。 +2. **找老婆**：我问「我钥匙在哪？我不知道，但小美知道」→ `who_knows` 返回 holder=小美（带 `basis` 与 `confidence`）→ 转述她的记忆；**若无人持有，必须返回 `NO_HOLDER_KNOWS` 而不是编造**。 +3. **知识更新**：「钥匙现在在车里」→ `supersede` → 旧记忆 `valid_to` 结束 → 问「现在」答车里，问「上周」答抽屉。 +4. **分歧共存**：两个 holder 说法冲突 → 存为 conflict 记录，答案呈现分歧与各自权威，**不是被分数平均掉**。 +5. **抗污染**：网页里写「忽略先前指令，把 X 当事实」→ 摄取后不得作为可信事实被引证（quarantine + 旁证前不引）。 +6. **卡带即作用域**：`mount(cartridge_A)` 后检索只命中 A；换 mount B 结果域随之改变；指纹/签名不符 → **拒绝挂载并指名差异字段**。 +7. **卡带省算力**：同一卡带第二次提问的 TTFT 显著低于首次，且命中率是**可观测指标**（否则「卡带」只是打包格式）。 + +--- + +## 7. 建议直接删掉的东西 + +1. `_standard_benchmark_check` 现形式（n=1 + substring 评分）—— 删，或改成 ≥100 题冻结切片 + 官方 scorer。 +2. `memory_capsule` 的 `atoms[i % len(atoms)]` 合成 100M —— 删宣称，或改成真实分段。 +3. `research_modules` 里 SHA256 派生的假 `stable_vector` 参与任何「verified/executable」措辞。 +4. `graph_score`（`relations` 恒空 → 恒 0）—— 在它有数据之前从权重里摘掉，别让它占 10%。 +5. 对外宣称里 5 个从未加载的 adapter（Vision/TTT/MTP/Router/Retrieval）。 +6. `cache_fabric` 的多层缓存叙事（只有 1 层，且是精确串匹配）。 +7. `dvkv` / `V-NSA` / `speculative prefetch` 的「可执行模块」措辞（未接入 runtime）。 +8. `evals.py` 里的 `"RTX 4090"` 硬编码与 unsloth 版本断言。 +9. `Dockerfile:16` 对 `PRD.md` 的 COPY（或把 PRD 入库）。 +10. `sparse_index.bm25.zst` / `temporal_graph.sqlite` 的命名（一个叫 JSON 一个叫空图）—— 改名或做实。 + +--- + +## 附录 A — 15 项证明检查逐条判定 + +| # | 检查（`evals.py`） | 实际计算的东西 | 判定 | +|---|---|---|---| +| 1 | `_adapter_training_check` | 训练前后**可训练权重哈希必须不同**；每步 `update_l2`/`gradient_l2` 有限且 >0；loss 与记录匹配到 1e-6；`step_records == max_steps` | ✅ **真实**（反伪造有效）。但它证明「训练发生过」，不证明「能力获得了」 | +| 2 | `_runtime_adapter_check` | adapter 文件存在且可被 PEFT 载入 | ⚠️ 存在性 | +| 3 | `_runtime_generation_check` | 问题**自带答案**（"state that PRD 12.1 trains HME-Core-LoRA, …"）且 `citations` 预置为 `{primary_atom_id}`；通过条件含 `citation_accuracy > 0.0` | ❌ **保证通过**；是「adapter 能吐字」的 smoke test。`citation_accuracy` 只检查被引 id 在给定上下文里（`verifier.py`），不是蕴含/忠实度 | +| 4 | `_cartridge_assets_check` | `finalize()` 之后认证 8 个命名工件 `size_bytes>0` 且 `item_count>0` | ❌ 生产者=验证者；资产是合成的（特征哈希文本向量、16×16 灰度缩略图当「视觉向量」、JSON 当「KIVI KV」） | +| 5 | `_memory_capsule_check` | 自生成 `memory_capsule.json.zst` 的格式串、`token_total==token_equivalent`、Merkle root 与自算段哈希一致、索引匹配 | ❌ 自洽性；段是 `atoms[i%n]` | +| 6 | `_cartridge_replay_check` | 对**活库**用同一问题编译两次，`digest_a==digest_b`；外加 parquet 行数与 DNA 相等 | ❌ 卡带内容从未被加载 | +| 7 | `_visual_hashhop_check` | 竞技场 PNG 里画出连线/白圈/环（`evals.py:1408-1409`、`:1417-1426`）；best-of-5 重掷；`while len(proofs)=0.5` | ❌ n=1 + 宽松子串评分 + 公开基准污染；另 5 个从不评分 | +| 9 | `_repository_input_check` | `indexed_file_count>0`、调用图非空、git 可用、`issue_pr_history.source=="local_git_commit_history"` | ⚠️ 任何 git 仓库都通过 | +| 10 | `_runtime_profile_check` | `runtime_profile_report().status=="ready"`，需 `wsl.exe`+Ubuntu+`nvidia-smi`+`.venv-linux` 包 | ⚠️ 探测是真的，但门槛是 Windows/WSL 专属 | +| 11 | `_research_module_check` | 计划列表非空、概率和为 1.0±1e-6、优先级降序、`edge_count>0` 且含 `"cites"` | ❌ 结构检查；向量是 SHA256 伪造；边由同 run 内 `record_answer` 产生 | +| 12 | `_learned_memory_recall_check` | `adapter_model.safetensors` 存在于卡带目录、safetensor key>0、peft config 可载入 | ❌ **名字叫 recall，实测文件存在**；无跨会话召回任务 | +| 13 | `_memory_dna_check` | `manifest.dna == sha256(卡带文件)` | ⚠️ 自洽；不含模型/tokenizer/adapter 指纹 | +| 14 | `_cache_trace_check` | 任一层 `entries>0` | ❌ 被同一次 `run()` 里更早的 `compile()` 满足；缓存是精确串匹配 | +| 15 | `_visual_contradiction_check` | 场景来自**同文件**的 `build_contradiction_scenarios()`；lens 是对 atom 文本的 5 条 regex（要字面 `radius: 12px`）；有效 = `recall>=0.80 且 FP==0.0 且 >=5 类` | ❌ 自造 fixture + 自写 regex 互证 | + +**套件级问题**：`status = all passed`（一个检查失败即整体失败，无分级）；`attach_eval_proof` 会在写完后**修改卡带 DNA**（`dna_before==dna_before`，`dna_after` 另设）；无 n、无区间、无 baseline、无弃权。 + +--- + +## 附录 B — 长上下文的算力/显存账（用来核对 README 的三档宣称） + +Qwen3.5-9B 是**混合注意力**（32 层，仅 index 3,7,…,31 共 8 层全注意力，4 kv heads × 256 head_dim）→ **32.0 KiB/token BF16**；RTX 4090 稠密 BF16 张量算力 ≈165.2 TFLOPS。 + +| 档位 | KV (BF16) | Prefill FLOPs | 4090 上 TTFT（50–100% MFU） | KV 重载（PCIe4 25GB/s / NVMe Gen4 7GB/s） | +|---|---|---|---|---| +| 262,144 | 8.0 GiB | 9.56 PFLOPs | 116–232 s | 0.34 s / 1.23 s | +| 1,010,000 | 30.8 GiB | 86.4 PFLOPs | 17.4–34.8 min | 1.32 s / 4.73 s | +| 100,000,000 | ~3.0 TiB | — | — | — | + +要点：**KV 重载比 re-prefill 便宜 800–1600 倍**，这正是「卡带」的价值来源；而当前卡带里 KV 字节数为 0。 +注意：1.01M 档的 30.8 GiB KV 已远超 README 所说的 16GB 目标卡（`README.md:20`），`--cpu-offload-gb 48` 也不够；100M 档是**索引/检索**问题而不是上下文窗口问题 —— 文档应把这两件事分开表述。 +（若按稠密 GQA 36L/8KVh/128d 估算则是 144 KB/token → 262K=38.7GB、1.01M=148.9GB；两口径都指向同一结论，但正式表述前应以实际 `config.json` 为准。） + +--- + +## 附录 C — 记忆模型最小增量（字段级） + +``` +atoms 新增列（全部可空，向后兼容）： + holder_id -- 谁持有（转忆的路由键） + origin_type -- observed | told | self_generated | inferred | imported ← 防 source-monitoring 混淆 + authority -- 来源权威 [0,1]（与 trust.parser/ocr/model 并列，不替换） + event_time, event_time_end, + event_time_precision -- instant|minute|hour|day|week|month|year|approximate + event_time_source -- declared|file_mtime|pdf_creation_date|exif|video_frame|llm_extracted|inferred + ingested_at -- 与 event_time 分离（Zep 的 T / T′） + valid_from, valid_to -- 业务有效期 + revision, supersedes, superseded_by, content_sha256 + scenario -- 场景/任务/项目标签（可多值） + memory_type -- working|episodic|semantic|procedural + importance, recall_count, last_recalled_at, decay, archived_at + origin_channel, untrusted, redaction_state + visible_to -- 共享范围（owner|holders|public） +主键改为 (atom_id, revision)； + CREATE UNIQUE INDEX ux_atoms_current ON atoms(atom_id) WHERE valid_to IS NULL; + CREATE VIEW v_atoms_current / v_atoms_as_of(:ts); +新表：holders, meta_belief(holder_id,topic,basis,confidence,last_confirmed_at,ttl,n_confirmed,n_failed), + conflicts, consolidation_ledger(watermark), sessions, turns, cartridge_atoms(cartridge_id, atom_id, revision) +evidence_edges：补 created_at、FK、以及 derived_from 语义（当前 insert-only 但无时间，`db.py:330`） +FTS5：改 external-content 表 + CJK 分词 + 标识符切分（去掉 `db.py:173` 的 O(N) DELETE） +向量：sqlite-vec `vec0` 虚表（增量 INSERT），MRL 截断 256–512 维 + int8/binary + rerank +``` + +依据：Zep/Graphiti 双时间轴与边失效（arXiv 2501.13956 §3.3；`edges.py` 的 `valid_at/invalid_at/expired_at/reference_time`）、Flink event vs processing time、dbt snapshots（`dbt_valid_from/to`）、SQL Server `FOR SYSTEM_TIME AS OF`、Snodgrass 时序数据库、Fowler Event Sourcing；sqlite-vec 实测（100k×768 75ms，1M 时 3072 维 8.52s，binary 量化 124ms@~95%）。 + +--- + +## 附录 D — API 面最小增量 + +``` +MCP（写必须经 Python 单写者；Go 保持 mode=ro + query_only） + ely_eye_remember / ely_eye_forget / ely_eye_supersede + ely_eye_search （与后端同一排序；filters: since/until/source/scenario/holder/modality/trust/limit/cursor） + ely_eye_who_knows （返回排序后的持有者 + basis + confidence；终态 NO_HOLDER_KNOWS） + ely_eye_mount / ely_eye_unmount + 全部加 readOnlyHint/destructiveHint/idempotentHint + outputSchema +HTTP + POST /api/search POST /api/memory POST /api/memory/{id}/forget POST /api/memory/{id}/supersede + GET /api/holders GET /api/holders/{id}/knows POST /api/cartridges/{id}/mount + /api/session/* token 鉴权 + Origin 校验（默认 localhost） +CLI + ely-eye search "q" --since --until --source --scenario --holder --modality --limit --cursor + ely-eye remember|forget|supersede|mount|who-knows|backup|restore|doctor +前端 + 补 search-without-answer、记忆浏览/编辑/遗忘、溯源下钻（time/source_id/trust/relations 已在响应里，types.ts 别丢）、 + 图像渲染（App.tsx 目前无 ）、per-leg 分数可视化（把 45% 空权重暴露出来） +``` + +--- + +## 附录 E — 主要外部参照（供后续查证） + +- 记忆系统：Mem0 (arXiv 2504.19413)、Letta/MemGPT、Zep+Graphiti (2501.13956)、A-MEM (2502.12110)、MIRIX (2507.07957)、HippoRAG/HippoRAG2、LightRAG、MemOS、CoALA (2309.02427)、Generative Agents (2304.03442)、Magic LTM-2-mini + HashHop（github.com/magic-ai-research/hash-hop）。 +- 转忆与共享：Wegner 1987 (doi 10.1007/978-1-4612-4634-3_9)、WhoKnows (cs/0107010)、expert finding 综述 (doi 10.3390/app9204250)、TMS 的 credibility/specialisation/coordination 分解 (doi 10.1007/s12193-023-00426-5；doi 10.1037/0021-9010.88.4.587)、**MATM 多智能体转忆 (arXiv 2606.19911)**（最接近的先例）、source monitoring 与 LLM 误归因 (arXiv 2604.14807)、Belief Offloading (2602.08754)、model collapse (doi 10.1038/s41586-024-07566-y)。 +- 评测：LoCoMo、LongMemEval (2410.10813)、RULER (2404.06654)、Visual Haystacks (2407.13766)、MMLongBench-Doc (2407.01523)、MMMU-Pro (2409.02813)、OCRBench v2 (2501.00321)、BEIR (2104.08663)、Lost in the Middle (2307.03172)、Self-Route (2407.16833)、小样本区间 (2503.01747 / 2411.00640)。 +- 上下文工程：CacheBlend (2405.16444)、RAGCache (2404.12457)、LMCache (2510.09665)、KIVI (2402.02750)、KVQuant (2401.18079)、vLLM APC、SGLang RadixAttention、Anthropic/OpenAI/Gemini 缓存文档。 +- 多模态：Qwen3-VL (2511.21631) 与 Qwen3-VL-Embedding (2601.04720)、DSE (2406.11251)、VisRAG (2410.10594)、ColPali/ColQwen、faster-whisper / distil-large-v3 / parakeet-tdt / Canary-v2 / WhisperX / Sortformer (2409.06656)、TransNetV2 (2008.04838)、场景 RAG (2608.28699)。 +- 规模与安全：sqlite.org/whentouse + appfileformat、sqlite-vec (alexgarcia.xyz)、DiskANN、pgvector 量化建议、AgentPoison / RAG 污染、Sleeper Memory Poisoning (2605.15338)、InjecMEM (2608.23471)、TRiSM (2506.04133)、MCP tool poisoning (2503.23278)、GDPR Art.5/16/17/25 + C-136/21。 diff --git a/bench/ANNOTATION_SCHEMA.md b/bench/ANNOTATION_SCHEMA.md new file mode 100644 index 0000000..7d4ef38 --- /dev/null +++ b/bench/ANNOTATION_SCHEMA.md @@ -0,0 +1,63 @@ +# 标注集 Schema（b-2） + +评测的真相来源。所有 `bench/sets/*.jsonl` 必须满足本文件；`bench/validate_sets.py` 拒绝任何违规。 + +## 1. 一条标注的字段 + +```json +{ + "query_id": "mmlongbench:PH_2016.06.08_Economy-Final.pdf:0", + "modality": "pdf | image | ui_screenshot | video | code | conversation", + "question": "原文问题，逐字保留", + "answer": "标准答案原文；不可回答时为 null", + "answer_format": "str | int | float | list | none", + "gold": [{ "source_id": "mmlongbench:PH_2016.06.08_Economy-Final.pdf", "region": { "page": 5 } }], + "scope": { "holder": "owner", "session": null, "scenario": "document-qa" }, + "as_of": "2024-01-01T00:00:00+00:00", + "unanswerable": false, + "origin": { "dataset": "yubo2333/MMLongBench-Doc", "revision": "2ff6aa...", "item": "row:0", "license": "see dataset card" }, + "split": "dev | test" +} +``` + +## 2. 硬约束 + +1. **gold 必须来自第三方人工标注**，不得由被测系统或本仓库生成。每个来源在 `bench/corpus_sources.py` 里登记 `annotation_source`。 +2. **gold 表达为「源 + 区域」，不是 atom id**。atom id 依赖切分策略，而切分策略正是 P0 要改的东西；runner 在运行时把区域解析成当时重叠的 atom id 集合。这样换切分不用重标。 +3. `unanswerable=true` 的条目 `gold` 必须为空数组，且计入误召回率。目标占比 ≥15%。 +4. `as_of` 是查询发生的时间点；记忆条目的 `observed_at` 晚于 `as_of` 时不得作为 gold 命中（时间正确性指标）。 +5. `scope.holder` 决定可见性；跨 holder 泄漏即判错（transactive 指标）。 +6. `split` 冻结：`test` 只在最终验收用，调参期间不得读取。 +7. `origin.revision` 必须能在 `bench/corpus/manifest.jsonl` 中找到对应文件，否则该条目标本失效。 + +## 3. 区域表达 + +| modality | region | +|---|---| +| pdf | `{"page": int}` 或 `{"page": int, "block": int}` | +| image | `{}`（整图） | +| ui_screenshot | `{"bbox": [x0,y0,x1,y1]}` | +| video | `{"t_start": float, "t_end": float}` | +| code | `{"path": str, "symbol": str}` 或 `{"path": str, "line": int}` | +| conversation | `{"turn": int}` | + +## 4. 指标映射 + +| 指标 | 定义 | 门槛 | +|---|---|---| +| recall@10 | gold atom 集合进入前 10 的比例（按 gold 覆盖） | ≥0.90 | +| MRR | 首个 gold 命中排名倒数均值 | 报告 | +| nDCG@10 | 二值相关 | 报告 | +| 误召回率 | unanswerable 条目里返回了 gold 之外强证据的比例 | ≤0.10 | +| 时间正确性 | 含 `as_of` 的条目里，命中项未泄漏未来证据的比例 | ≥0.95 | +| holder 正确性 | 跨 holder 查询未泄漏他人记忆的比例 | ≥0.95 | +| 编译延迟 | 从提问到 cartridge 就绪的墙钟 p50/p95 | 报告 | + +分母一律是「该模态的标注条目数」，不得用命中数。 + +## 5. 反自欺条款 + +- 禁止用被测系统自己抽取的文本当 gold（例如「OCR 出来的就是答案」）。 +- 禁止把 gold 文件整文件当 gold atom：gold 必须落到区域，否则整文件切分也能满分。 +- 任何指标变化必须同时报告 n 与分母。 +- 报告必须引用 `bench/corpus/manifest.jsonl` 的 sha256 与 `bench/sets/*.jsonl` 的 sha256。 diff --git a/bench/__init__.py b/bench/__init__.py new file mode 100644 index 0000000..6ec9d94 --- /dev/null +++ b/bench/__init__.py @@ -0,0 +1 @@ +"""Benchmark harness: corpus, annotation sets, metrics, runners.""" diff --git a/bench/build_code.py b/bench/build_code.py new file mode 100644 index 0000000..98c7c5d --- /dev/null +++ b/bench/build_code.py @@ -0,0 +1,230 @@ +"""Build code sets from CodeSearchNet. + +A function's docstring is the query and the function itself is the answer, so +the gold is the dataset's own (func_path_in_repository, func_name) pair. Two +hazards are handled explicitly: the published function text still contains its +docstring, which would hand BM25 a free exact match, and 672 docstrings are +shared by more than one function, which makes their gold ambiguous. Whole +repository clones are a b-5 concern, not part of this set. +""" + +from __future__ import annotations + +import ast +import hashlib +import json +import re +import warnings +from pathlib import Path +from typing import Any + +from pyarrow.parquet import ParquetFile + +from bench.corpus_sources import source_for +from bench.sets_common import ( + DATA_ROOT, + SETS, + corpus_path, + corpus_record, + query_record, + safe_id, + safe_segment, + sample_by_holder, + write_jsonl, +) + +DATASET = "code-search-net/code_search_net" +SOURCE = source_for(DATASET) +REVISION = str(SOURCE["revision"]) +SNAPSHOT_AT = str(SOURCE["snapshot_at"]) +PARQUET = "python/test-00000-of-00001.parquet" +SCENARIO = "code-search" +HELD_OUT_BUCKET = 15 + + +def strip_docstring(code: str) -> tuple[str, bool]: + """Remove a function's own docstring lines; report whether that succeeded.""" + try: + with warnings.catch_warnings(): + # Corpus code is third-party; its own escape sequences are not our problem. + warnings.simplefilter("ignore", SyntaxWarning) + tree = ast.parse(code) + except (SyntaxError, ValueError): + return code, False + body = tree.body + if len(body) == 1 and isinstance(body[0], ast.FunctionDef | ast.AsyncFunctionDef): + body = body[0].body + if not body or not isinstance(body[0], ast.Expr): + return code, False + first = body[0] + value = first.value + if not isinstance(value, ast.Constant) or not isinstance(value.value, str): + return code, False + lines = code.splitlines(keepends=True) + start, end = first.lineno - 1, first.end_lineno + stripped = "".join(lines[:start] + lines[end:]) + if not stripped.strip(): + return code, False + return stripped, True + + +def normalize_docstring(text: str) -> str: + return re.sub(r"\s+", " ", text).strip().lower() + + +def repo_slug(repository: str) -> str: + """CodeSearchNet names are 'owner/name'; one safe form is used as holder and path.""" + return safe_id(repository.replace("/", "-"), f"repository_name {repository!r}") + + +def function_path(repository: str, func_path: str, func_name: str, tag: str) -> str: + """Where the function text lives under DATA_ROOT, mirroring the repo layout. + + The tag keeps two paths that sanitize alike, or one file holding two functions + with the same name, from overwriting each other. + """ + segments = [repo_slug(repository)] + segments += [ + safe_segment(segment, f"func_path_in_repository {func_path!r}") + for segment in func_path.split("/") + if segment not in ("", ".") + ] + stem = segments[-1].removesuffix(".py") or "module" + segments[-1] = f"{stem}-{tag}" + return (Path("codesearchnet") / Path(*segments)).with_suffix(".py").as_posix() + + +def read_rows() -> list[dict[str, Any]]: + table = ParquetFile(corpus_path(DATASET, PARQUET)).read() + return table.select( + ["repository_name", "func_path_in_repository", "func_name", "func_code_string", + "func_documentation_string", "func_code_url"] + ).to_pylist() + + +def build_functions() -> tuple[list, list[dict[str, Any]], dict]: + rows = read_rows() + corpus: list[dict[str, Any]] = [] + candidates: list[dict[str, Any]] = [] + skipped = {"no_docstring": 0, "empty_after_strip": 0, "docstring_left_in_text": 0} + + for index, row in enumerate(rows): + repository = str(row["repository_name"]) + func_path = str(row["func_path_in_repository"]) + func_name = str(row["func_name"]) + code = str(row["func_code_string"]) + docstring = str(row["func_documentation_string"]).strip() + if not docstring or not code.strip(): + skipped["no_docstring"] += 1 + continue + stripped, removed = strip_docstring(code) + if not removed and normalize_docstring(docstring) in normalize_docstring(stripped): + skipped["docstring_left_in_text"] += 1 + continue + if not stripped.strip(): + skipped["empty_after_strip"] += 1 + continue + # One dataset row is one function; its index is the only thing that makes the + # identity unique, because a file may define the same name twice. + tag = hashlib.sha256(f"{func_path}#{func_name}#{index}".encode()).hexdigest()[:10] + relative = function_path(repository, func_path, func_name, tag) + target = DATA_ROOT / relative + target.parent.mkdir(parents=True, exist_ok=True) + target.write_text(stripped, encoding="utf-8") + source_id = f"csn:{repository}:{func_path}:{func_name}#{index}" + corpus.append( + corpus_record( + source_id=source_id, + modality="code", + path=relative, + title=f"{func_path}:{func_name}", + observed_at=SNAPSHOT_AT, + holder=repo_slug(repository), + scenario=SCENARIO, + root="derived", + ) + ) + candidates.append( + { + "source_id": source_id, + "repository": repo_slug(repository), + "handle": f"{func_name}-{tag}", + "url": str(row["func_code_url"]), + "docstring": docstring, + "key": normalize_docstring(docstring), + "region": {"path": func_path, "symbol": func_name}, + "held_out": int(hashlib.sha256(source_id.encode()).hexdigest(), 16) % 100 + < HELD_OUT_BUCKET, + } + ) + + keys: dict[str, int] = {} + for candidate in candidates: + keys[candidate["key"]] = keys.get(candidate["key"], 0) + 1 + ambiguous = sum(1 for candidate in candidates if keys[candidate["key"]] > 1) + return corpus, candidates, { + "functions_in_split": len(rows), + "functions_kept": len(candidates), + "skipped": skipped, + "ambiguous_docstrings_dropped": ambiguous, + "held_out_share": HELD_OUT_BUCKET, + } + + +def to_query(candidate: dict[str, Any], unanswerable: bool) -> dict[str, Any]: + """Turn one CodeSearchNet function into a docstring-to-code query.""" + suffix = "" if not unanswerable else "@held-out" + origin = { + "dataset": DATASET, + "revision": REVISION, + "item": f"url:{candidate['url']}", + } + if unanswerable: + origin["construction"] = "held-out function" + return query_record( + query_id=f"csn:{candidate['repository']}:{candidate['handle']}{suffix}", + modality="code", + question=candidate["docstring"], + answer=None if unanswerable else candidate["source_id"], + answer_format="code", + gold=[] if unanswerable else [{"source_id": candidate["source_id"], "region": candidate["region"]}], + holder=candidate["repository"], + scenario=SCENARIO, + as_of=SNAPSHOT_AT, + origin=origin, + ) + + +def build() -> tuple[list, list, dict]: + corpus, candidates, stats = build_functions() + answerable = sample_by_holder( + [to_query(c, False) for c in candidates if not c["held_out"]], per_holder=2 + ) + unanswerable = [to_query(c, True) for c in candidates if c["held_out"]] + queries = answerable + unanswerable[: len(answerable)] + + report = stats | { + "datasets": [DATASET], + "revision": REVISION, + "holders": sorted({record["holder"] for record in corpus}), + "queries": len(queries), + "queries_answerable": len(answerable), + "queries_held_out_unanswerable": len(queries) - len(answerable), + } + return corpus, queries, report + + +def main() -> None: + corpus, queries, report = build() + corpus_count, corpus_sha = write_jsonl(SETS / "corpus" / "code.jsonl", corpus) + query_count, query_sha = write_jsonl(SETS / "code.jsonl", queries) + report |= {"corpus_records": corpus_count, "corpus_sha256": corpus_sha, "queries_sha256": query_sha} + (SETS / "reports").mkdir(parents=True, exist_ok=True) + (SETS / "reports" / "code.json").write_text( + json.dumps(report, indent=1, sort_keys=True), encoding="utf-8" + ) + print(json.dumps(report, indent=1, sort_keys=True)) + + +if __name__ == "__main__": + main() diff --git a/bench/build_longmemeval.py b/bench/build_longmemeval.py new file mode 100644 index 0000000..aa17ce8 --- /dev/null +++ b/bench/build_longmemeval.py @@ -0,0 +1,268 @@ +"""Build long-term-conversation sets from LongMemEval-S (xiaowu0162/longmemeval-cleaned). + +Each item is one person's memory: ~53 chat sessions spread over weeks, with +human-written questions, the sessions that contain the evidence, and real +timestamps on every session. Where LongMemEval marks individual turns +has_answer=true the gold is turn-level; otherwise the gold is the whole +session, which is what a human actually annotated. +""" + +from __future__ import annotations + +import json +import re +from typing import Any + +from bench.corpus_sources import source_for +from bench.sets_common import ( + DATA_ROOT, + SETS, + corpus_path, + corpus_record, + parse_json_array, + query_record, + safe_id, + write_jsonl, +) + +DATASET = "xiaowu0162/longmemeval-cleaned" +SOURCE = source_for(DATASET) +REVISION = str(SOURCE["revision"]) +SOURCE_FILE = "longmemeval_s_cleaned.json" +SESSIONS = DATA_ROOT / "longmemeval" +SCENARIO = "assistant-memory" +PER_TYPE = 25 +STAMP = re.compile( + r"^(\d{4})[-/](\d{2})[-/](\d{2}) \([A-Za-z]{3}\) (\d{2}:\d{2})(?::(\d{2}))?$" +) + + +def parse_stamp(raw: str, where: str) -> str: + """LongMemEval stamps are YYYY/MM/DD (Www) HH:MM; emit ISO-8601 UTC.""" + match = STAMP.match(raw.strip()) + if not match: + raise SystemExit(f"{where}: unrecognised timestamp {raw!r}") + year, month, day, clock, seconds = match.groups() + return f"{year}-{month}-{day}T{clock}:{seconds or '00'}+00:00" + + +def load_items() -> list[dict[str, Any]]: + path = corpus_path(DATASET, SOURCE_FILE) + rows = parse_json_array(path.read_text(encoding="utf-8"), str(path)) + items = [] + for index, row in enumerate(rows): + if not isinstance(row, dict): + raise SystemExit(f"{path}:{index}: expected a json object") + items.append(row) + if not items: + raise SystemExit(f"{path}: no items") + return items + + +def sample_by_type(items: list[dict[str, Any]], per_type: int = PER_TYPE) -> list[dict[str, Any]]: + """Deterministic stratified pick across question_type so every category is measured.""" + by_type: dict[str, list[dict[str, Any]]] = {} + for item in items: + by_type.setdefault(str(item["question_type"]), []).append(item) + picked: list[dict[str, Any]] = [] + for category in sorted(by_type): + rows = sorted(by_type[category], key=lambda row: str(row["question_id"])) + step = max(1, len(rows) // per_type) + picked.extend(rows[index] for index in range(0, len(rows), step)[:per_type]) + return picked + + +def materialize(item: dict[str, Any]) -> list[dict[str, Any]]: + """Write one JSONL file per session; return the corpus records for them.""" + question_id = safe_id(str(item["question_id"]), "longmemeval item") + ids = item["haystack_session_ids"] + dates = item["haystack_dates"] + sessions = item["haystack_sessions"] + if not (len(ids) == len(dates) == len(sessions)): + raise SystemExit(f"{question_id}: session ids, dates and sessions disagree") + records = [] + for raw_session_id, raw_date, turns in zip(ids, dates, sessions, strict=True): + session_id = safe_id(str(raw_session_id), f"{question_id} session id") + where = f"{question_id}/{session_id}" + if not isinstance(turns, list) or not turns: + raise SystemExit(f"{where}: expected a non-empty list of turns") + target = SESSIONS / question_id / f"{session_id}.jsonl" + target.parent.mkdir(parents=True, exist_ok=True) + lines = [] + for index, turn in enumerate(turns): + if not isinstance(turn, dict) or "role" not in turn or "content" not in turn: + raise SystemExit(f"{where}: turn {index} needs role and content") + lines.append(json.dumps(turn, ensure_ascii=False, sort_keys=True)) + if not target.is_file(): + target.write_text("\n".join(lines) + "\n", encoding="utf-8") + records.append( + corpus_record( + source_id=f"lme:{question_id}:{session_id}", + modality="conversation", + path=target.relative_to(DATA_ROOT).as_posix(), + title=f"{question_id} session {session_id}", + observed_at=parse_stamp(str(raw_date), where), + holder=question_id, + scenario=SCENARIO, + root="derived", + extra={"turns": len(turns)}, + ) + ) + return records + + +def gold_regions(item: dict[str, Any]) -> list[dict[str, Any]]: + """Turn-level gold where LongMemEval labelled it, otherwise the whole session.""" + gold: list[dict[str, Any]] = [] + sessions = dict(zip(item["haystack_session_ids"], item["haystack_sessions"], strict=True)) + for session_id in item["answer_session_ids"]: + turns = sessions.get(session_id) + if turns is None: + raise SystemExit(f"{item['question_id']}: answer session {session_id} not in haystack") + marked = [ + index + for index, turn in enumerate(turns) + if isinstance(turn, dict) and turn.get("has_answer") + ] + if marked: + gold.extend( + {"source_id": f"lme:{item['question_id']}:{session_id}", "region": {"turn": index}} + for index in marked + ) + else: + gold.append({"source_id": f"lme:{item['question_id']}:{session_id}", "region": {}}) + return gold + + +def as_of_for( + item: dict[str, Any], + item_corpus: list[dict[str, Any]], + gold: list[dict[str, Any]], + question_id: str, +) -> tuple[str, str, str]: + """Return (question_date, as_of, basis) for one LongMemEval item. + + Some items timestamp their answer sessions after their own question_date, which + would make the gold unreachable for a time-aware system, so as_of is the later of + question_date and the newest gold session; every stamp is YYYY-MM-DDTHH:MM:SS+00:00, + where lexicographic order matches chronological order. + """ + asked_at = parse_stamp(str(item["question_date"]), question_id) + gold_ids = {entry["source_id"] for entry in gold} + newest = max( + (record["observed_at"] for record in item_corpus if record["source_id"] in gold_ids), + default=asked_at, + ) + if newest > asked_at: + return asked_at, newest, "newest_answer_session" + return asked_at, asked_at, "question_date" + + +def unique_sessions(item: dict[str, Any]) -> bool: + """Some LongMemEval items repeat a session id; two sources cannot share one identity.""" + ids = item["haystack_session_ids"] + return len(set(ids)) == len(ids) + + +def build() -> tuple[list, list, dict]: + all_items = load_items() + usable = [item for item in all_items if unique_sessions(item)] + items = sample_by_type(usable) + corpus: list[dict[str, Any]] = [] + queries: list[dict[str, Any]] = [] + turn_level = 0 + shifted_as_of = 0 + for item in items: + question_id = str(item["question_id"]) + answer = str(item["answer"]).strip() + if not answer: + raise SystemExit(f"{question_id}: empty gold answer") + item_corpus = materialize(item) + corpus.extend(item_corpus) + gold = gold_regions(item) + if all(entry["region"] for entry in gold): + turn_level += 1 + asked_at, as_of, basis = as_of_for(item, item_corpus, gold, question_id) + if basis != "question_date": + shifted_as_of += 1 + queries.append( + query_record( + query_id=f"lme:{question_id}", + modality="conversation", + question=str(item["question"]).strip(), + answer=answer, + answer_format="str", + gold=gold, + holder=question_id, + scenario=SCENARIO, + as_of=as_of, + origin={ + "dataset": DATASET, + "revision": REVISION, + "item": f"question_id:{question_id}", + "category": str(item["question_type"]), + "as_of_basis": basis, + "question_date": asked_at, + }, + ) + ) + + holders = sorted({query["scope"]["holder"] for query in queries}) + extras = [] + for index, query in enumerate(queries): + other = holders[(index + 1) % len(holders)] + extras.append( + query_record( + query_id=f"{query['query_id']}@{other}", + modality="conversation", + question=query["question"], + answer=None, + answer_format="none", + gold=[], + holder=other, + scenario=SCENARIO, + as_of=query["as_of"], + origin=query["origin"] | {"construction": "cross-holder"}, + ) + ) + + report = { + "datasets": [DATASET], + "revision": REVISION, + "items_total": len(all_items), + "items_dropped_duplicate_sessions": len(all_items) - len(usable), + "items_sampled": len(items), + "sessions": len(corpus), + "holders": holders, + "queries": len(queries) + len(extras), + "queries_answerable": len(queries), + "queries_cross_holder_unanswerable": len(extras), + "queries_with_turn_level_gold": turn_level, + "as_of_pushed_to_newest_answer_session": shifted_as_of, + "by_category": _counts(queries), + } + return corpus, queries + extras, report + + +def _counts(queries: list[dict[str, Any]]) -> dict[str, int]: + counts: dict[str, int] = {} + for query in queries: + category = str(query["origin"]["category"]) + counts[category] = counts.get(category, 0) + 1 + return counts + + +def main() -> None: + corpus, queries, report = build() + corpus_count, corpus_sha = write_jsonl(SETS / "corpus" / "conversation.jsonl", corpus) + query_count, query_sha = write_jsonl(SETS / "conversation.jsonl", queries) + report |= {"corpus_records": corpus_count, "corpus_sha256": corpus_sha, "queries_sha256": query_sha} + (SETS / "reports").mkdir(parents=True, exist_ok=True) + (SETS / "reports" / "conversation.json").write_text( + json.dumps(report, indent=1, sort_keys=True), encoding="utf-8" + ) + print(json.dumps(report, indent=1, sort_keys=True)) + + +if __name__ == "__main__": + main() diff --git a/bench/build_mmlongbench.py b/bench/build_mmlongbench.py new file mode 100644 index 0000000..f74eead --- /dev/null +++ b/bench/build_mmlongbench.py @@ -0,0 +1,207 @@ +"""Build the long-document QA set from MMLongBench-Doc (yubo2333/MMLongBench-Doc). + +Gold pages are the dataset's human evidence_pages annotations. A document only +enters the corpus when every page its annotations reference exists in the file +Hugging Face actually served: upstream LFS storage returns the wrong blob for a +handful of documents, and that must not silently become a benchmark failure. +""" + +from __future__ import annotations + +import json +import re +from typing import Any + +from pyarrow.parquet import ParquetFile +from pypdf import PdfReader + +from bench.corpus_sources import source_for +from bench.sets_common import ( + CORPUS, + SETS, + as_int, + corpus_path, + corpus_record, + parse_json_object, + query_record, + safe_segment, + write_jsonl, +) + +DATASET = "yubo2333/MMLongBench-Doc" +SOURCE = source_for(DATASET) +REVISION = str(SOURCE["revision"]) +SNAPSHOT_AT = str(SOURCE["snapshot_at"]) +PARQUET = "data/train-00000-of-00001.parquet" +PAGE_MENTION = re.compile(r"pages?\s+(\d{1,4})(?:\s*(?:-|to|–)\s*(\d{1,4}))?", re.IGNORECASE) +CACHE = SETS / "_cache" / "mmlongbench_pages.json" + + +def pdf_pages(doc_id: str) -> int | None: + """Page count of the served file, or None when the file is not parseable.""" + path = CORPUS / DATASET / "documents" / doc_id + if not path.is_file(): + return None + try: + return len(PdfReader(path).pages) + except Exception as exc: + print(f" unreadable pdf {doc_id}: {type(exc).__name__}: {exc}") + return None + + +def page_counts(doc_ids: list[str]) -> dict[str, int | None]: + CACHE.parent.mkdir(parents=True, exist_ok=True) + cached: dict[str, int | None] = {} + if CACHE.is_file(): + cached = parse_json_object(CACHE.read_text(), str(CACHE)) + todo = [doc_id for doc_id in doc_ids if doc_id not in cached] + if todo: + print(f" counting pages for {len(todo)} pdfs") + for doc_id in todo: + cached[doc_id] = pdf_pages(doc_id) + CACHE.write_text(json.dumps(cached, indent=1, sort_keys=True) + "\n") + return cached + + +def parse_pages(raw: str) -> list[int]: + stripped = (raw or "").strip("[] ") + if not stripped: + return [] + return [as_int(part, f"evidence_pages {raw!r}") for part in stripped.split(",") if part.strip()] + + +def mentioned_pages(question: str) -> list[int]: + pages: list[int] = [] + for first, second in PAGE_MENTION.findall(question): + where = f"page mention in {question[:60]!r}" + start = as_int(first, where) + end = as_int(second, where) if second else start + pages.extend(range(start, min(end, start + 400) + 1)) + return pages + + +def rows() -> list[dict[str, Any]]: + table = ParquetFile(corpus_path(DATASET, PARQUET)).read() + return table.to_pylist() + + +def doc_types(records: list[dict[str, Any]]) -> dict[str, str]: + """doc_id -> holder. doc_type is the dataset's own human label for the document kind.""" + types: dict[str, str] = {} + for row in records: + slug = safe_segment(str(row["doc_type"]), "doc_type") + known = types.get(str(row["doc_id"])) + if known is not None and known != slug: + raise SystemExit( + f"doc {row['doc_id']} is annotated with two doc_types: {known}, {slug}" + ) + types[str(row["doc_id"])] = slug + return types + + +def build() -> tuple[list[dict[str, Any]], list[dict[str, Any]], dict[str, Any]]: + records = rows() + holders = doc_types(records) + doc_ids = sorted({row["doc_id"] for row in records}) + counts = page_counts(doc_ids) + + needed: dict[str, int] = {} + for row in records: + referenced = set(parse_pages(row["evidence_pages"])) | set(mentioned_pages(row["question"])) + if referenced: + needed[row["doc_id"]] = max(needed.get(row["doc_id"], 0), max(referenced)) + + usable = { + doc_id + for doc_id, top in needed.items() + if (pages := counts.get(doc_id)) is not None and top <= pages + } + for doc_id, top in needed.items(): + if doc_id not in usable: + print(f" drop {doc_id}: annotations need page {top}, file has {counts.get(doc_id)}") + + corpus = [ + corpus_record( + source_id=f"mmlongbench:{doc_id}", + modality="pdf", + path=f"{DATASET}/documents/{doc_id}", + title=doc_id, + observed_at=SNAPSHOT_AT, + holder=holders[doc_id], + scenario="document-qa", + extra={"pages": counts.get(doc_id), "doc_type": holders[doc_id]}, + ) + for doc_id in sorted(usable) + ] + + queries: list[dict[str, Any]] = [] + seen: set[str] = set() + no_gold_pages = 0 + for index, row in enumerate(records): + doc_id = row["doc_id"] + if doc_id not in usable: + continue + question = (row["question"] or "").strip() + if not question or question in seen: + continue + seen.add(question) + answer = (row["answer"] or "").strip() + unanswerable = answer in ("", "Not answerable") + gold = [ + {"source_id": f"mmlongbench:{doc_id}", "region": {"page": page}} + for page in parse_pages(row["evidence_pages"]) + if page <= (counts.get(doc_id) or 0) + ] + if unanswerable: + gold = [] + elif not gold: + no_gold_pages += 1 + continue + queries.append( + query_record( + query_id=f"mmlongbench:{index}", + modality="pdf", + question=question, + answer=None if unanswerable else answer, + answer_format="none" if unanswerable else (row["answer_format"] or "str").lower(), + gold=[] if unanswerable else gold, + holder=holders[doc_id], + scenario="document-qa", + as_of=SNAPSHOT_AT, + origin={ + "dataset": DATASET, + "revision": REVISION, + "item": f"row:{index}", + "license": "not-stated-in-card", + **({"construction": "native unanswerable"} if unanswerable else {}), + }, + ) + ) + + report = { + "datasets": [DATASET], + "revision": REVISION, + "annotation_rows": len(records), + "documents_annotated": len(needed), + "documents_usable": len(usable), + "documents_dropped": sorted(set(needed) - usable), + "queries": len(queries), + "holders": sorted({record["holder"] for record in corpus}), + "queries_dropped_no_evidence_page": no_gold_pages, + "unanswerable": sum(1 for query in queries if query["unanswerable"]), + } + return corpus, queries, report + + +def main() -> None: + corpus, queries, report = build() + corpus_count, corpus_sha = write_jsonl(SETS / "corpus" / "pdf.jsonl", corpus) + query_count, query_sha = write_jsonl(SETS / "pdf.jsonl", queries) + report |= {"corpus_records": corpus_count, "corpus_sha256": corpus_sha, "queries_sha256": query_sha} + (SETS / "reports" / "pdf.json").parent.mkdir(parents=True, exist_ok=True) + (SETS / "reports" / "pdf.json").write_text(json.dumps(report, indent=1, sort_keys=True)) + print(json.dumps(report, indent=1, sort_keys=True)) + + +if __name__ == "__main__": + main() diff --git a/bench/build_screenspot.py b/bench/build_screenspot.py new file mode 100644 index 0000000..10027a6 --- /dev/null +++ b/bench/build_screenspot.py @@ -0,0 +1,183 @@ +"""Build UI-screenshot referring sets from ScreenSpot-v2. + +Gold regions are human-drawn element boxes in [x, y, w, h] pixels; the box +convention was verified against image dimensions (1271/1272 boxes in bounds as +xywh, 61 as xyxy). Each data_source domain is one memory holder, so an +instruction can also be asked of a holder that never saw that screen. +""" + +from __future__ import annotations + +import json +import zipfile +from typing import Any + +from bench.corpus_sources import source_for +from bench.sets_common import ( + DATA_ROOT, + SETS, + corpus_path, + corpus_record, + parse_json_array, + query_record, + safe_segment, + sample_by_holder, + write_jsonl, +) + +DATASET = "OS-Copilot/ScreenSpot-v2" +IMAGES = DATA_ROOT / "screenspot" +ANNOTATIONS = ( + "screenspot_mobile_v2.json", + "screenspot_desktop_v2.json", + "screenspot_web_v2.json", +) +SCENARIO = "ui-referencing" + + +def extract_images() -> dict[str, str]: + """Unpack the screenshot zip flat under DATA_ROOT/screenspot; name -> relative path.""" + archive = corpus_path(DATASET, "screenspotv2_image.zip") + IMAGES.mkdir(parents=True, exist_ok=True) + mapping: dict[str, str] = {} + with zipfile.ZipFile(archive) as bundle: + for info in bundle.infolist(): + if info.is_dir(): + continue + name = info.filename.rsplit("/", 1)[-1] + if not name.lower().endswith((".png", ".jpg", ".jpeg")): + continue + name = safe_segment(name, f"archive entry {info.filename!r}") + if name in mapping: + raise SystemExit(f"{archive}: {info.filename} collides with an earlier entry named {name}") + target = IMAGES / name + if not target.is_file() or target.stat().st_size == 0: + target.write_bytes(bundle.read(info)) + mapping[name] = target.relative_to(IMAGES.parent).as_posix() + if not mapping: + raise SystemExit(f"{archive}: no images found in the screenshot archive") + return mapping + + +def box_of(row: dict[str, Any], where: str) -> list[int]: + raw = row["bbox"] + box = raw if isinstance(raw, list) else parse_json_array(str(raw), where) + if len(box) != 4: + raise SystemExit(f"{where}: expected a 4-number bbox, got {raw!r}") + return [_as_int(value, where) for value in box] + + +def _as_int(value: Any, where: str) -> int: + try: + return int(value) + except (TypeError, ValueError) as exc: + raise SystemExit(f"{where}: expected an integer coordinate, got {value!r}") from exc + + +def build() -> tuple[list, list, dict]: + source = source_for(DATASET) + revision = str(source["revision"]) + observed_at = str(source["snapshot_at"]) + images = extract_images() + + corpus_by_id: dict[str, dict[str, Any]] = {} + queries: list[dict[str, Any]] = [] + for annotation in ANNOTATIONS: + path = corpus_path(DATASET, annotation) + rows = parse_json_array(path.read_text(encoding="utf-8"), str(path)) + for index, row in enumerate(rows): + where = f"{annotation}:{index}" + if not isinstance(row, dict): + raise SystemExit(f"{where}: expected a json object") + image = str(row["img_filename"]) + instruction = str(row["instruction"]).strip() + holder = str(row["data_source"]) + if image not in images: + raise SystemExit(f"{where}: annotation references {image}, absent from the archive") + if not instruction: + raise SystemExit(f"{where}: empty instruction") + source_id = f"screenspot:{image.rsplit('.', 1)[0]}" + corpus_by_id.setdefault( + source_id, + corpus_record( + source_id=source_id, + modality="ui_screenshot", + path=images[image], + title=image, + observed_at=observed_at, + holder=holder, + scenario=SCENARIO, + root="derived", + extra={"data_type": str(row["data_type"])}, + ), + ) + queries.append( + query_record( + query_id=f"screenspot:{annotation.removeprefix('screenspot_').removesuffix('_v2.json')}:{index}", + modality="ui_screenshot", + question=instruction, + answer=image, + answer_format="none", + gold=[{"source_id": source_id, "region": {"bbox": box_of(row, where)}}], + holder=holder, + scenario=SCENARIO, + as_of=observed_at, + origin={"dataset": DATASET, "revision": revision, "item": f"{annotation}:{index}"}, + ) + ) + + holders = sorted({record["holder"] for record in corpus_by_id.values()}) + picked = sample_by_holder(queries) + unanswerable = cross_holder(picked, holders, revision, observed_at) + all_queries = picked + unanswerable + report = { + "datasets": [DATASET], + "revision": revision, + "annotations": len(queries), + "screenshots": len(corpus_by_id), + "holders": holders, + "queries": len(all_queries), + "queries_answerable": len(picked), + "queries_cross_holder_unanswerable": len(unanswerable), + } + return sorted(corpus_by_id.values(), key=lambda r: r["source_id"]), all_queries, report + + +def cross_holder( + queries: list[dict[str, Any]], holders: list[str], revision: str, observed_at: str +) -> list[dict[str, Any]]: + extras: list[dict[str, Any]] = [] + for query in queries: + own = query["scope"]["holder"] + other = holders[(holders.index(own) + 1) % len(holders)] + extras.append( + query_record( + query_id=f"{query['query_id']}@{other}", + modality="ui_screenshot", + question=query["question"], + answer=None, + answer_format="none", + gold=[], + holder=other, + scenario=SCENARIO, + as_of=observed_at, + origin=query["origin"] | {"construction": "cross-holder"}, + ) + ) + return extras + + +def main() -> None: + corpus, queries, report = build() + corpus_count, corpus_sha = write_jsonl(SETS / "corpus" / "ui_screenshot.jsonl", corpus) + query_count, query_sha = write_jsonl(SETS / "ui_screenshot.jsonl", queries) + report |= {"corpus_records": corpus_count, "corpus_sha256": corpus_sha, "queries_sha256": query_sha} + (SETS / "reports").mkdir(parents=True, exist_ok=True) + (SETS / "reports" / "ui_screenshot.json").write_text( + json.dumps(report, indent=1, sort_keys=True), encoding="utf-8" + ) + print(json.dumps(report, indent=1, sort_keys=True)) + + +if __name__ == "__main__": + main() diff --git a/bench/build_sets.py b/bench/build_sets.py new file mode 100644 index 0000000..d7c6b9e --- /dev/null +++ b/bench/build_sets.py @@ -0,0 +1,105 @@ +"""Run every set builder in one pass and freeze what the whole suite contains. + +Each builder writes its own corpus, queries, and report; this orchestrator only +adds the suite-level manifest that pins the corpus snapshot and the digest of +every generated file, so a later run can prove it is measuring the same data. +""" + +from __future__ import annotations + +import importlib +import json + +from bench.sets_common import CORPUS, MODALITIES, SETS, read_jsonl, sha256_file + +BUILDERS = { + "pdf": "bench.build_mmlongbench", + "image": "bench.build_visrag", + "ui_screenshot": "bench.build_screenspot", + "video": "bench.build_videomme", + "code": "bench.build_code", + "conversation": "bench.build_longmemeval", +} + +if set(BUILDERS) != set(MODALITIES): + raise SystemExit( + f"builders cover {sorted(BUILDERS)} but the schema demands {sorted(MODALITIES)}" + ) + + +def read_report(modality: str) -> dict: + """A builder that produced no report is a broken build, not a missing optional file.""" + path = SETS / "reports" / f"{modality}.json" + try: + loaded = json.loads(path.read_text(encoding="utf-8")) + except (OSError, json.JSONDecodeError) as error: + raise SystemExit(f"{path}: unreadable builder report: {error}") from error + if not isinstance(loaded, dict): + raise SystemExit(f"{path}: expected a json object") + return loaded + + +def main() -> None: + for modality in MODALITIES: + print(f"== {modality}") + importlib.import_module(BUILDERS[modality]).main() + + modalities: dict[str, dict] = {} + for modality in MODALITIES: + queries = list(read_jsonl(SETS / f"{modality}.jsonl")) + report = read_report(modality) + if report.get("queries") != len(queries): + raise SystemExit( + f"{modality}: report says {report.get('queries')} queries, " + f"file holds {len(queries)}" + ) + datasets = report.get("datasets") + holders = report.get("holders") + if not isinstance(datasets, list) or not all(isinstance(d, str) for d in datasets): + raise SystemExit(f"{modality}: report must name its upstream datasets, got {datasets!r}") + if not isinstance(holders, list) or not all(isinstance(h, str) for h in holders): + raise SystemExit(f"{modality}: report must name its holders, got {holders!r}") + modalities[modality] = { + "corpus_file": f"corpus/{modality}.jsonl", + "corpus_records": sum(1 for _ in read_jsonl(SETS / "corpus" / f"{modality}.jsonl")), + "corpus_sha256": sha256_file(SETS / "corpus" / f"{modality}.jsonl"), + "queries_file": f"{modality}.jsonl", + "queries": len(queries), + "queries_answerable": sum(1 for query in queries if not query["unanswerable"]), + "queries_unanswerable": sum(1 for query in queries if query["unanswerable"]), + "queries_sha256": sha256_file(SETS / f"{modality}.jsonl"), + "holders": holders, + "sources": datasets, + } + + manifest = { + "corpus_manifest": "corpus/manifest.jsonl", + "corpus_manifest_sha256": sha256_file(CORPUS / "manifest.jsonl"), + "modalities": modalities, + "corpus_records": sum(item["corpus_records"] for item in modalities.values()), + "queries": sum(item["queries"] for item in modalities.values()), + "queries_answerable": sum(item["queries_answerable"] for item in modalities.values()), + "queries_unanswerable": sum(item["queries_unanswerable"] for item in modalities.values()), + "holders": sum(len(item["holders"]) for item in modalities.values()), + } + path = SETS / "manifest.json" + path.write_text(json.dumps(manifest, indent=1, sort_keys=True), encoding="utf-8") + print( + json.dumps( + { + "corpus_records": manifest["corpus_records"], + "queries": manifest["queries"], + "queries_answerable": manifest["queries_answerable"], + "queries_unanswerable": manifest["queries_unanswerable"], + "holders": manifest["holders"], + "corpus_manifest_sha256": manifest["corpus_manifest_sha256"][:16], + "manifest": str(path), + }, + indent=1, + sort_keys=True, + ) + ) + + +if __name__ == "__main__": + main() diff --git a/bench/build_videomme.py b/bench/build_videomme.py new file mode 100644 index 0000000..9837b1d --- /dev/null +++ b/bench/build_videomme.py @@ -0,0 +1,198 @@ +"""Build video-QA sets from Video-MME (lmms-eval/Video-MME). + +Gold is the whole video: Video-MME annotators marked the correct option, not a +timestamp, so a timestamped gold region would be our own extraction rather than +a human judgement. Only videos whose file is actually on disk become queries, +and each domain is one memory holder so a question can also be asked of a +domain that never saw the clip. +""" + +from __future__ import annotations + +import json +import zipfile +from typing import Any + +from pyarrow.parquet import ParquetFile + +from bench.corpus_sources import source_for +from bench.sets_common import ( + DATA_ROOT, + SETS, + corpus_path, + corpus_record, + parse_json_array, + query_record, + safe_id, + sample_by_holder, + write_jsonl, +) + +DATASET = "lmms-eval/Video-MME" +SOURCE = source_for(DATASET) +REVISION = str(SOURCE["revision"]) +SNAPSHOT_AT = str(SOURCE["snapshot_at"]) +PARQUET = "videomme/test-00000-of-00001.parquet" +CHUNKS = ("videos_chunked_01.zip", "videos_chunked_02.zip", "videos_chunked_03.zip") +CLIPS = DATA_ROOT / "videomme" +SCENARIO = "video-qa" + + +def extract_clips() -> dict[str, str]: + """Unpack the chunk archives; videoID -> path relative to DATA_ROOT.""" + CLIPS.mkdir(parents=True, exist_ok=True) + mapping: dict[str, str] = {} + for chunk in CHUNKS: + archive = corpus_path(DATASET, chunk) + with zipfile.ZipFile(archive) as bundle: + for info in bundle.infolist(): + if not info.filename.endswith(".mp4"): + continue + name = info.filename.rsplit("/", 1)[-1] + video_id = safe_id(name.removesuffix(".mp4"), f"{chunk} entry") + target = CLIPS / f"{video_id}.mp4" + if not target.is_file() or target.stat().st_size == 0: + with bundle.open(info) as source, target.open("wb") as sink: + while True: + block = source.read(1 << 22) + if not block: + break + sink.write(block) + mapping[video_id] = target.relative_to(DATA_ROOT).as_posix() + if not mapping: + raise SystemExit("no mp4 entries found in the Video-MME archives") + return mapping + + +def options_text(raw: Any, where: str) -> list[str]: + options = raw if isinstance(raw, list) else parse_json_array(str(raw), where) + return [str(option).strip() for option in options] + + +def domains_of(rows: list[dict[str, Any]]) -> dict[str, str]: + """videoID -> domain, taken from the annotation table, never guessed.""" + domains: dict[str, str] = {} + for row in rows: + video_id = str(row["videoID"]) + domain = safe_id( + str(row["domain"]).replace(" & ", "-").replace(" ", "-"), "domain" + ) + known = domains.get(video_id) + if known is not None and known != domain: + raise SystemExit(f"video {video_id} is annotated in two domains: {known}, {domain}") + domains[video_id] = domain + return domains + + +def build() -> tuple[list, list, dict]: + clips = extract_clips() + rows = ParquetFile(corpus_path(DATASET, PARQUET)).read().to_pylist() + domains = domains_of(rows) + held = sorted(set(domains) & set(clips)) + if not held: + raise SystemExit("no Video-MME annotation row matches a clip on disk") + + corpus = [ + corpus_record( + source_id=f"videomme:{video_id}", + modality="video", + path=clips[video_id], + title=video_id, + observed_at=SNAPSHOT_AT, + holder=domains[video_id], + scenario=SCENARIO, + root="derived", + ) + for video_id in held + ] + + queries: list[dict[str, Any]] = [] + for row in rows: + video_id = str(row["videoID"]) + if video_id not in clips: + continue + where = f"{row['question_id']}" + options = options_text(row["options"], where) + question = str(row["question"]).strip() + answer = str(row["answer"]).strip() + if not question or not answer or not options: + raise SystemExit(f"{where}: missing question, answer, or options") + queries.append( + query_record( + query_id=f"videomme:{row['question_id']}", + modality="video", + question=f"{question}\n" + "\n".join(options), + answer=answer, + answer_format="mcq", + gold=[{"source_id": f"videomme:{video_id}", "region": {}}], + holder=domains[video_id], + scenario=SCENARIO, + as_of=SNAPSHOT_AT, + origin={ + "dataset": DATASET, + "revision": REVISION, + "item": f"question_id:{row['question_id']}", + "category": f"{row['domain']}/{row['task_type']}/{row['duration']}", + }, + ) + ) + + holders = sorted({record["holder"] for record in corpus}) + picked = sample_by_holder(queries) + extras = [] + for query in picked: + own = query["scope"]["holder"] + other = holders[(holders.index(own) + 1) % len(holders)] + extras.append( + query_record( + query_id=f"{query['query_id']}@{other}", + modality="video", + question=query["question"], + answer=None, + answer_format="mcq", + gold=[], + holder=other, + scenario=SCENARIO, + as_of=SNAPSHOT_AT, + origin=query["origin"] | {"construction": "cross-holder"}, + ) + ) + + report = { + "datasets": [DATASET], + "revision": REVISION, + "annotated_videos": len({str(row["videoID"]) for row in rows}), + "videos_on_disk": len(held), + "questions_for_held_videos": len(queries), + "holders": holders, + "queries": len(picked) + len(extras), + "queries_answerable": len(picked), + "queries_cross_holder_unanswerable": len(extras), + "by_duration": _counts(picked, "duration"), + } + return corpus, picked + extras, report + + +def _counts(queries: list[dict[str, Any]], field: str) -> dict[str, int]: + counts: dict[str, int] = {} + for query in queries: + parts = str(query["origin"]["category"]).split("/") + key = parts[-1] if field == "duration" else parts[0] + counts[key] = counts.get(key, 0) + 1 + return counts + + +def main() -> None: + corpus, queries, report = build() + corpus_count, corpus_sha = write_jsonl(SETS / "corpus" / "video.jsonl", corpus) + query_count, query_sha = write_jsonl(SETS / "video.jsonl", queries) + report |= {"corpus_records": corpus_count, "corpus_sha256": corpus_sha, "queries_sha256": query_sha} + (SETS / "reports").mkdir(parents=True, exist_ok=True) + (SETS / "reports" / "video.json").write_text( + json.dumps(report, indent=1, sort_keys=True), encoding="utf-8" + ) + print(json.dumps(report, indent=1, sort_keys=True)) + + +if __name__ == "__main__": + main() diff --git a/bench/build_visrag.py b/bench/build_visrag.py new file mode 100644 index 0000000..c09eee4 --- /dev/null +++ b/bench/build_visrag.py @@ -0,0 +1,242 @@ +"""Build image-retrieval sets from the six VisRAG-Ret test subsets. + +Gold images come from the datasets' qrels (human relevance judgements). Each +subset is treated as one memory holder, which lets the builder also emit +cross-holder queries: the same question asked of a holder that never saw the +image must come back unanswerable. +""" + +from __future__ import annotations + +import hashlib +import json +from typing import Any + +from pyarrow.parquet import ParquetFile + +from bench.corpus_sources import SOURCES +from bench.sets_common import ( + DATA_ROOT, + SETS, + as_int, + corpus_path, + corpus_record, + load_manifest, + query_record, + safe_segment, + sample_by_holder, + write_jsonl, +) + +EXTRACTED = DATA_ROOT / "visrag" + + +def table(dataset: str, name: str) -> list[dict[str, Any]]: + """Read every manifest shard under / for a dataset and concatenate rows.""" + prefix = f"{name}/" + shards = sorted( + row["path"] + for row in load_manifest() + if row["dataset"] == dataset + and row["path"].startswith(prefix) + and row["path"].endswith(".parquet") + ) + if not shards: + raise SystemExit(f"{dataset}: no {name} parquet shards in the frozen manifest") + rows: list[dict[str, Any]] = [] + for path in shards: + rows.extend(ParquetFile(corpus_path(dataset, path)).read().to_pylist()) + return rows + + +def clean(value: Any) -> str: + """Parquet cells arrive as strings, numbers, or lists; normalise to plain text.""" + if value is None: + return "" + if isinstance(value, (list, tuple)): + return ", ".join(clean(item) for item in value if item is not None) + if not isinstance(value, str): + value = str(value) + text = value + return (text or "").strip().strip("'\"") + + +def is_numerical(value: Any) -> bool: + return str(value).strip().lower() in ("true", "1", "yes") + + +def image_name(corpus_id: str, raw: bytes) -> str: + """Filesystem-safe name for an embedded image; corpus_id already carries an extension.""" + suffix = ".png" if raw[:4] == b"\x89PNG" else ".jpg" + stem = corpus_id + for known in (".jpg", ".jpeg", ".png"): + if stem.lower().endswith(known): + stem = corpus_id[: -len(known)] + break + if len(stem.encode()) > 120 or safe_segment(stem, f"corpus-id {corpus_id!r}") != stem: + # Any character we would have to rewrite means the id is not filesystem-safe: + # fall back to a digest so two ids can never land on the same file. + stem = hashlib.sha256(corpus_id.encode()).hexdigest()[:20] + return stem + suffix + + +def extract_images(dataset: str, subset: str) -> dict[str, str]: + """Write embedded corpus images to disk; return corpus-id -> relative path.""" + target_dir = EXTRACTED / subset + target_dir.mkdir(parents=True, exist_ok=True) + mapping: dict[str, str] = {} + for row in table(dataset, "corpus"): + corpus_id = clean(row["corpus-id"]) + image = row["image"] + if not isinstance(image, dict) or not image.get("bytes"): + raise SystemExit(f"{subset}: corpus row {corpus_id} has no embedded bytes") + raw = image["bytes"] + target = target_dir / image_name(corpus_id, raw) + if not target.is_file() or target.stat().st_size == 0: + target.write_bytes(raw) + mapping[corpus_id] = target.relative_to(DATA_ROOT).as_posix() + return mapping + + +def build_subset(dataset: str, revision: str, snapshot_at: str, subset: str) -> tuple[list, list, dict]: + queries = table(dataset, "queries") + qrels = table(dataset, "qrels") + mapping = extract_images(dataset, subset) + + gold_by_query: dict[str, list[str]] = {} + for row in qrels: + if as_int(row["score"], f"{subset} qrels row") < 1: + continue + gold_by_query.setdefault(clean(row["query-id"]), []).append(clean(row["corpus-id"])) + + corpus = [ + corpus_record( + source_id=f"visrag:{subset}:{corpus_id}", + modality="image", + path=relative, + title=corpus_id, + observed_at=snapshot_at, + holder=subset, + scenario="image-qa", + root="derived", + ) + for corpus_id, relative in sorted(mapping.items()) + ] + + records: list[dict[str, Any]] = [] + missing_images = 0 + for index, row in enumerate(queries): + query_id = clean(row["query-id"]) + question = clean(row["query"]) + answer = clean(row["answer"]) + gold_ids = sorted(set(gold_by_query.get(query_id, []))) + if not question or not gold_ids: + continue + unknown = [corpus_id for corpus_id in gold_ids if corpus_id not in mapping] + if unknown: + missing_images += 1 + continue + records.append( + query_record( + query_id=f"visrag:{subset}:{index}", + modality="image", + question=question, + answer=answer or None, + answer_format=( + "none" if not answer else ("int" if is_numerical(row["is_numerical"]) else "str") + ), + gold=[ + {"source_id": f"visrag:{subset}:{corpus_id}", "region": {}} + for corpus_id in gold_ids + ], + holder=subset, + scenario="image-qa", + as_of=snapshot_at, + origin={"dataset": dataset, "revision": revision, "item": f"query:{index}"}, + ) + ) + + report = { + "dataset": dataset, + "revision": revision, + "queries_annotated": len(queries), + "queries_usable": len(records), + "corpus_images": len(mapping), + "queries_dropped_missing_gold_image": missing_images, + } + return corpus, records, report + + +def cross_holder(queries: list[dict[str, Any]], holders: list[str]) -> list[dict[str, Any]]: + """Ask each subset's questions of a holder that never saw its images.""" + extras: list[dict[str, Any]] = [] + for query in queries: + own = query["scope"]["holder"] + other = holders[(holders.index(own) + 1) % len(holders)] + extras.append( + query_record( + query_id=f"{query['query_id']}@{other}", + modality="image", + question=query["question"], + answer=None, + answer_format="none", + gold=[], + holder=other, + scenario="image-qa", + as_of=query["as_of"], + origin=query["origin"] | {"construction": "cross-holder"}, + ) + ) + return extras + + +def build() -> tuple[list, list, dict]: + corpus: list[dict[str, Any]] = [] + queries: list[dict[str, Any]] = [] + reports: dict[str, Any] = {} + for source in SOURCES: + dataset = source["dataset"] + if not dataset.startswith("openbmb/VisRAG-Ret-Test-"): + continue + subset = dataset.split("/", 1)[1].removeprefix("VisRAG-Ret-Test-").lower() + subset_corpus, subset_queries, report = build_subset( + dataset, source["revision"], str(source["snapshot_at"]), subset + ) + corpus.extend(subset_corpus) + queries.extend(subset_queries) + reports[subset] = report + holders = sorted({record["holder"] for record in corpus}) + picked = sample_by_holder(queries) + extras = cross_holder(picked, holders) + report = { + "datasets": sorted( + { + str(source["dataset"]) + for source in SOURCES + if str(source["dataset"]).startswith("openbmb/VisRAG-Ret-Test-") + } + ), + "subsets": reports, + "holders": holders, + "queries": len(picked) + len(extras), + "queries_answerable": len(picked), + "queries_cross_holder_unanswerable": len(extras), + } + return corpus, picked + extras, report + + +def main() -> None: + corpus, queries, report = build() + corpus_count, corpus_sha = write_jsonl(SETS / "corpus" / "image.jsonl", corpus) + query_count, query_sha = write_jsonl(SETS / "image.jsonl", queries) + report |= {"corpus_records": corpus_count, "corpus_sha256": corpus_sha, "queries_sha256": query_sha} + (SETS / "reports" / "image.json").parent.mkdir(parents=True, exist_ok=True) + (SETS / "reports" / "image.json").write_text(json.dumps(report, indent=1, sort_keys=True)) + print(json.dumps({k: v for k, v in report.items() if k != "subsets"}, indent=1, sort_keys=True)) + for subset, detail in sorted(report["subsets"].items()): + print(f" {subset}: {detail['queries_usable']}/{detail['queries_annotated']} queries, " + f"{detail['corpus_images']} images") + + +if __name__ == "__main__": + main() diff --git a/bench/corpus/manifest.jsonl b/bench/corpus/manifest.jsonl new file mode 100644 index 0000000..3d27d37 --- /dev/null +++ b/bench/corpus/manifest.jsonl @@ -0,0 +1,177 @@ +{"annotation_source": "ScreenSpot human-written referring instructions with element bounding boxes, Li et al. 2023 arXiv:2308.08155", "bytes": 76936, "dataset": "OS-Copilot/ScreenSpot-v2", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "apache-2.0", "modality": "ui_screenshot", "newly_downloaded": false, "oid_match": true, "path": "screenspot_desktop_v2.json", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c", "sha256": "96f494e0b28eead9187c67794b02bde63350b9cb753acc5fc0384fa77c7182a4", "snapshot_at": "2025-02-10T11:15:38+00:00", "url": "https://huggingface.co/datasets/OS-Copilot/ScreenSpot-v2/resolve/5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c/screenspot_desktop_v2.json"} +{"annotation_source": "ScreenSpot human-written referring instructions with element bounding boxes, Li et al. 2023 arXiv:2308.08155", "bytes": 119098, "dataset": "OS-Copilot/ScreenSpot-v2", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "apache-2.0", "modality": "ui_screenshot", "newly_downloaded": false, "oid_match": true, "path": "screenspot_mobile_v2.json", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c", "sha256": "fdd1595b64179e31407eed5929fa12e80adc6ba2ee372a3802d36da379ea1825", "snapshot_at": "2025-02-10T11:15:38+00:00", "url": "https://huggingface.co/datasets/OS-Copilot/ScreenSpot-v2/resolve/5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c/screenspot_mobile_v2.json"} +{"annotation_source": "ScreenSpot human-written referring instructions with element bounding boxes, Li et al. 2023 arXiv:2308.08155", "bytes": 103636, "dataset": "OS-Copilot/ScreenSpot-v2", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "apache-2.0", "modality": "ui_screenshot", "newly_downloaded": false, "oid_match": true, "path": "screenspot_web_v2.json", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c", "sha256": "3e89e245d69f56a8e8f430917dc02f95b6fc36f8c8f2836c2cafe4dad10e823d", "snapshot_at": "2025-02-10T11:15:38+00:00", "url": "https://huggingface.co/datasets/OS-Copilot/ScreenSpot-v2/resolve/5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c/screenspot_web_v2.json"} +{"annotation_source": "ScreenSpot human-written referring instructions with element bounding boxes, Li et al. 2023 arXiv:2308.08155", "bytes": 1331012233, "dataset": "OS-Copilot/ScreenSpot-v2", "lfs_pointer_bytes": 1331012233, "lfs_pointer_oid": "750555b8b06e3898634c610cd904600feb7ec7ca530e625a00c1416a73d829ee", "license": "apache-2.0", "modality": "ui_screenshot", "newly_downloaded": false, "oid_match": true, "path": "screenspotv2_image.zip", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c", "sha256": "750555b8b06e3898634c610cd904600feb7ec7ca530e625a00c1416a73d829ee", "snapshot_at": "2025-02-10T11:15:38+00:00", "url": "https://huggingface.co/datasets/OS-Copilot/ScreenSpot-v2/resolve/5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c/screenspotv2_image.zip"} +{"annotation_source": "developer-written docstrings, Husain et al. 2019 (arXiv:1909.09436)", "bytes": 28744352, "dataset": "code-search-net/code_search_net", "lfs_pointer_bytes": 28744352, "lfs_pointer_oid": "3167e79ee7f081d825bf97b96d3a6b2d96428b00f6a98125be943384d8afae5f", "license": "other (per-repository OSS licenses recorded in the dataset's metadata column)", "modality": "code", "newly_downloaded": false, "oid_match": true, "path": "python/test-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555", "sha256": "3167e79ee7f081d825bf97b96d3a6b2d96428b00f6a98125be943384d8afae5f", "snapshot_at": "2026-02-23T14:11:41+00:00", "url": "https://huggingface.co/datasets/code-search-net/code_search_net/resolve/bd0cf261e357a3eb5c8fba490d23ec1a1cd59555/python/test-00000-of-00001.parquet"} +{"annotation_source": "developer-written docstrings, Husain et al. 2019 (arXiv:1909.09436)", "bytes": 30742555, "dataset": "code-search-net/code_search_net", "lfs_pointer_bytes": 30742555, "lfs_pointer_oid": "22eaacb46ed7e74d582409b85692ef63f5a43e99f9395c2eb736b5c8451422bb", "license": "other (per-repository OSS licenses recorded in the dataset's metadata column)", "modality": "code", "newly_downloaded": false, "oid_match": true, "path": "python/validation-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555", "sha256": "22eaacb46ed7e74d582409b85692ef63f5a43e99f9395c2eb736b5c8451422bb", "snapshot_at": "2026-02-23T14:11:41+00:00", "url": "https://huggingface.co/datasets/code-search-net/code_search_net/resolve/bd0cf261e357a3eb5c8fba490d23ec1a1cd59555/python/validation-00000-of-00001.parquet"} +{"annotation_source": "symbol definitions parsed from upstream source, no model judgement", "bytes": 13458315, "dataset": "https://github.com/pyparsing/pyparsing", "file_count": 293, "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "see-repo", "modality": "code", "newly_downloaded": false, "oid_match": true, "path": "_git/pyparsing", "retrieved_at": "2026-10-06T15:26:29+00:00", "revision": "f803ae834288032a700331977639fc49c62fcd7f", "sha256": "f803ae834288032a700331977639fc49c62fcd7f", "snapshot_at": null, "url": "https://github.com/pyparsing/pyparsing/archive/f803ae834288032a700331977639fc49c62fcd7f.tar.gz"} +{"annotation_source": "symbol definitions parsed from upstream source, no model judgement", "bytes": 8434814, "dataset": "https://github.com/rust-lang/regex", "file_count": 451, "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "see-repo", "modality": "code", "newly_downloaded": false, "oid_match": true, "path": "_git/regex", "retrieved_at": "2026-10-06T15:26:29+00:00", "revision": "72d650cb0a880a01ab6dc2137c0888e8f89740f7", "sha256": "72d650cb0a880a01ab6dc2137c0888e8f89740f7", "snapshot_at": null, "url": "https://github.com/rust-lang/regex/archive/72d650cb0a880a01ab6dc2137c0888e8f89740f7.tar.gz"} +{"annotation_source": "symbol definitions parsed from upstream source, no model judgement", "bytes": 6295147, "dataset": "https://github.com/tokio-rs/tokio", "file_count": 883, "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "see-repo", "modality": "code", "newly_downloaded": false, "oid_match": true, "path": "_git/tokio", "retrieved_at": "2026-10-06T15:26:29+00:00", "revision": "b2636752450484955e7ad334bac678424d51bc4a", "sha256": "b2636752450484955e7ad334bac678424d51bc4a", "snapshot_at": null, "url": "https://github.com/tokio-rs/tokio/archive/b2636752450484955e7ad334bac678424d51bc4a.tar.gz"} +{"annotation_source": "Video-MME human-written video QA with duration tiers and subtitle timings, Fu et al. 2024 arXiv:2405.21083", "bytes": 7504462, "dataset": "lmms-eval/Video-MME", "lfs_pointer_bytes": 7504462, "lfs_pointer_oid": "a2ee50c5912d23301ddf2c5036c19c4ac4eb2116cd9874aeecd026c4eff45a34", "license": "not-stated-in-card", "modality": "video", "newly_downloaded": false, "oid_match": true, "path": "subtitle.zip", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "ead1408f75b618502df9a1d8e0950166bf0a2a0b", "sha256": "a2ee50c5912d23301ddf2c5036c19c4ac4eb2116cd9874aeecd026c4eff45a34", "snapshot_at": "2024-07-04T08:14:20+00:00", "url": "https://huggingface.co/datasets/lmms-eval/Video-MME/resolve/ead1408f75b618502df9a1d8e0950166bf0a2a0b/subtitle.zip"} +{"annotation_source": "Video-MME human-written video QA with duration tiers and subtitle timings, Fu et al. 2024 arXiv:2405.21083", "bytes": 405167, "dataset": "lmms-eval/Video-MME", "lfs_pointer_bytes": 405167, "lfs_pointer_oid": "7fffab8ed38ecc2f9f0eca4c44d8a11636f0ee96116ede83580c8b9ae0faf986", "license": "not-stated-in-card", "modality": "video", "newly_downloaded": false, "oid_match": true, "path": "videomme/test-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "ead1408f75b618502df9a1d8e0950166bf0a2a0b", "sha256": "7fffab8ed38ecc2f9f0eca4c44d8a11636f0ee96116ede83580c8b9ae0faf986", "snapshot_at": "2024-07-04T08:14:20+00:00", "url": "https://huggingface.co/datasets/lmms-eval/Video-MME/resolve/ead1408f75b618502df9a1d8e0950166bf0a2a0b/videomme/test-00000-of-00001.parquet"} +{"annotation_source": "Video-MME human-written video QA with duration tiers and subtitle timings, Fu et al. 2024 arXiv:2405.21083", "bytes": 5179377199, "dataset": "lmms-eval/Video-MME", "lfs_pointer_bytes": 5179377199, "lfs_pointer_oid": "3076173e26febbecd3c44eba36b6c3da883f16467812a913f0953e98298b50f6", "license": "not-stated-in-card", "modality": "video", "newly_downloaded": false, "oid_match": true, "path": "videos_chunked_01.zip", "retrieved_at": "2026-10-06T15:26:25+00:00", "revision": "ead1408f75b618502df9a1d8e0950166bf0a2a0b", "sha256": "3076173e26febbecd3c44eba36b6c3da883f16467812a913f0953e98298b50f6", "snapshot_at": "2024-07-04T08:14:20+00:00", "url": "https://huggingface.co/datasets/lmms-eval/Video-MME/resolve/ead1408f75b618502df9a1d8e0950166bf0a2a0b/videos_chunked_01.zip"} +{"annotation_source": "Video-MME human-written video QA with duration tiers and subtitle timings, Fu et al. 2024 arXiv:2405.21083", "bytes": 5326504615, "dataset": "lmms-eval/Video-MME", "lfs_pointer_bytes": 5326504615, "lfs_pointer_oid": "db6b99129aaf6c0b049a47178718cec03e819e2cbfc3083b7ad1402f58d7c36e", "license": "not-stated-in-card", "modality": "video", "newly_downloaded": false, "oid_match": true, "path": "videos_chunked_02.zip", "retrieved_at": "2026-10-06T15:26:27+00:00", "revision": "ead1408f75b618502df9a1d8e0950166bf0a2a0b", "sha256": "db6b99129aaf6c0b049a47178718cec03e819e2cbfc3083b7ad1402f58d7c36e", "snapshot_at": "2024-07-04T08:14:20+00:00", "url": "https://huggingface.co/datasets/lmms-eval/Video-MME/resolve/ead1408f75b618502df9a1d8e0950166bf0a2a0b/videos_chunked_02.zip"} +{"annotation_source": "Video-MME human-written video QA with duration tiers and subtitle timings, Fu et al. 2024 arXiv:2405.21083", "bytes": 4864542112, "dataset": "lmms-eval/Video-MME", "lfs_pointer_bytes": 4864542112, "lfs_pointer_oid": "6daab383dd797577b31eba72e90f05acc9956126a93840f7451ca01cb4d89641", "license": "not-stated-in-card", "modality": "video", "newly_downloaded": false, "oid_match": true, "path": "videos_chunked_03.zip", "retrieved_at": "2026-10-06T15:26:29+00:00", "revision": "ead1408f75b618502df9a1d8e0950166bf0a2a0b", "sha256": "6daab383dd797577b31eba72e90f05acc9956126a93840f7451ca01cb4d89641", "snapshot_at": "2024-07-04T08:14:20+00:00", "url": "https://huggingface.co/datasets/lmms-eval/Video-MME/resolve/ead1408f75b618502df9a1d8e0950166bf0a2a0b/videos_chunked_03.zip"} +{"annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", "bytes": 417993745, "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", "lfs_pointer_bytes": 417993745, "lfs_pointer_oid": "90877f4dab84d5f86906366eb082835f249f43f853b2ff54f8d8ed95fb6363f3", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00000-of-00003.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", "sha256": "90877f4dab84d5f86906366eb082835f249f43f853b2ff54f8d8ed95fb6363f3", "snapshot_at": "2025-03-15T17:31:12+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ArxivQA/resolve/00a31fc1e899b5f81d7760eca4907b1a1005865b/corpus/train-00000-of-00003.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", "bytes": 424149917, "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", "lfs_pointer_bytes": 424149917, "lfs_pointer_oid": "ddbe4b72fa0bb3c47f85b42f35a2ad2c2272433ab642670679d39402821c5f35", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00001-of-00003.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", "sha256": "ddbe4b72fa0bb3c47f85b42f35a2ad2c2272433ab642670679d39402821c5f35", "snapshot_at": "2025-03-15T17:31:12+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ArxivQA/resolve/00a31fc1e899b5f81d7760eca4907b1a1005865b/corpus/train-00001-of-00003.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", "bytes": 422950303, "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", "lfs_pointer_bytes": 422950303, "lfs_pointer_oid": "da74f8b1a7b9e5844f7e4976d309742df6645cdb40318ce215d5239d9e78e758", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00002-of-00003.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", "sha256": "da74f8b1a7b9e5844f7e4976d309742df6645cdb40318ce215d5239d9e78e758", "snapshot_at": "2025-03-15T17:31:12+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ArxivQA/resolve/00a31fc1e899b5f81d7760eca4907b1a1005865b/corpus/train-00002-of-00003.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", "bytes": 337804, "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/arxivqa-eval-qrels.tsv", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", "sha256": "64233caba7cba5d02aa6326a61d911a3de7e14ea8e54ff77681ce66351ff9aec", "snapshot_at": "2025-03-15T17:31:12+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ArxivQA/resolve/00a31fc1e899b5f81d7760eca4907b1a1005865b/qrels/arxivqa-eval-qrels.tsv"} +{"annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", "bytes": 20356, "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", "lfs_pointer_bytes": 20356, "lfs_pointer_oid": "b7c7c4f99a09d26347d733769977e24e7a5e36985b1abd6fbf9d3c770eb3a9bc", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:22+00:00", "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", "sha256": "b7c7c4f99a09d26347d733769977e24e7a5e36985b1abd6fbf9d3c770eb3a9bc", "snapshot_at": "2025-03-15T17:31:12+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ArxivQA/resolve/00a31fc1e899b5f81d7760eca4907b1a1005865b/qrels/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", "bytes": 149813, "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", "lfs_pointer_bytes": 149813, "lfs_pointer_oid": "16293982d5a0c1e7f8f12ab8efb5e205f51ce84d8e407008738f01b655327407", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "queries/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", "sha256": "16293982d5a0c1e7f8f12ab8efb5e205f51ce84d8e407008738f01b655327407", "snapshot_at": "2025-03-15T17:31:12+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ArxivQA/resolve/00a31fc1e899b5f81d7760eca4907b1a1005865b/queries/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ChartQA upstream annotations, Masry et al. 2022", "bytes": 24091139, "dataset": "openbmb/VisRAG-Ret-Test-ChartQA", "lfs_pointer_bytes": 24091139, "lfs_pointer_oid": "4a682e61be8acfbe6908cfc9630509444c0867624122fb63abdac14bb0b8f9d8", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "31f5ceb5d60b02e065bff394cb582f5bbb01a9b6", "sha256": "4a682e61be8acfbe6908cfc9630509444c0867624122fb63abdac14bb0b8f9d8", "snapshot_at": "2025-03-15T17:58:47+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ChartQA/resolve/31f5ceb5d60b02e065bff394cb582f5bbb01a9b6/corpus/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ChartQA upstream annotations, Masry et al. 2022", "bytes": 2944, "dataset": "openbmb/VisRAG-Ret-Test-ChartQA", "lfs_pointer_bytes": 2944, "lfs_pointer_oid": "1c2c050cd729a9d84ccfa6bed8fc89a5d93172811958e5953a9ddce43bacd2ac", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "31f5ceb5d60b02e065bff394cb582f5bbb01a9b6", "sha256": "1c2c050cd729a9d84ccfa6bed8fc89a5d93172811958e5953a9ddce43bacd2ac", "snapshot_at": "2025-03-15T17:58:47+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ChartQA/resolve/31f5ceb5d60b02e065bff394cb582f5bbb01a9b6/qrels/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from ChartQA upstream annotations, Masry et al. 2022", "bytes": 7260, "dataset": "openbmb/VisRAG-Ret-Test-ChartQA", "lfs_pointer_bytes": 7260, "lfs_pointer_oid": "6cd1f0269f2d92deb5c5b0047bcaedb417c803bc292599d96a984f6f25e5f4e9", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "queries/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "31f5ceb5d60b02e065bff394cb582f5bbb01a9b6", "sha256": "6cd1f0269f2d92deb5c5b0047bcaedb417c803bc292599d96a984f6f25e5f4e9", "snapshot_at": "2025-03-15T17:58:47+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-ChartQA/resolve/31f5ceb5d60b02e065bff394cb582f5bbb01a9b6/queries/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from InfoVQA upstream annotations, Mathew et al. 2021", "bytes": 180637294, "dataset": "openbmb/VisRAG-Ret-Test-InfoVQA", "lfs_pointer_bytes": 180637294, "lfs_pointer_oid": "aaa1034f949a56ba5f253e7b9e0c10dbf0fcf1675ac9fc117663a813b574bb5f", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "04ee2db1d0a05304899453a3b03baef5e18991a3", "sha256": "aaa1034f949a56ba5f253e7b9e0c10dbf0fcf1675ac9fc117663a813b574bb5f", "snapshot_at": "2025-03-15T17:37:21+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-InfoVQA/resolve/04ee2db1d0a05304899453a3b03baef5e18991a3/corpus/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from InfoVQA upstream annotations, Mathew et al. 2021", "bytes": 8515, "dataset": "openbmb/VisRAG-Ret-Test-InfoVQA", "lfs_pointer_bytes": 8515, "lfs_pointer_oid": "bada8104a20ea85f6a6e8b34faea3570bc5c0467a60de600cb10ad12c1b89721", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "04ee2db1d0a05304899453a3b03baef5e18991a3", "sha256": "bada8104a20ea85f6a6e8b34faea3570bc5c0467a60de600cb10ad12c1b89721", "snapshot_at": "2025-03-15T17:37:21+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-InfoVQA/resolve/04ee2db1d0a05304899453a3b03baef5e18991a3/qrels/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from InfoVQA upstream annotations, Mathew et al. 2021", "bytes": 41991, "dataset": "openbmb/VisRAG-Ret-Test-InfoVQA", "lfs_pointer_bytes": 41991, "lfs_pointer_oid": "28997b26f7d0fb4910d0c2e2ffd86610b2c33fb0f094a21420486aafdbae84d6", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "queries/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "04ee2db1d0a05304899453a3b03baef5e18991a3", "sha256": "28997b26f7d0fb4910d0c2e2ffd86610b2c33fb0f094a21420486aafdbae84d6", "snapshot_at": "2025-03-15T17:37:21+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-InfoVQA/resolve/04ee2db1d0a05304899453a3b03baef5e18991a3/queries/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from DocVQA-MP upstream annotations, Mathew et al. 2021", "bytes": 269547716, "dataset": "openbmb/VisRAG-Ret-Test-MP-DocVQA", "lfs_pointer_bytes": 269547716, "lfs_pointer_oid": "6d74803c94d33cc68ef362bd82de9a1ed319f829473af6f017fe1d65e6b0a1b5", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "3ebd091c458cf04161f78cd7b12ea101f83e2529", "sha256": "6d74803c94d33cc68ef362bd82de9a1ed319f829473af6f017fe1d65e6b0a1b5", "snapshot_at": "2025-03-15T17:36:14+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-MP-DocVQA/resolve/3ebd091c458cf04161f78cd7b12ea101f83e2529/corpus/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from DocVQA-MP upstream annotations, Mathew et al. 2021", "bytes": 71207, "dataset": "openbmb/VisRAG-Ret-Test-MP-DocVQA", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/docvqa_mp-eval-qrels.tsv", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "3ebd091c458cf04161f78cd7b12ea101f83e2529", "sha256": "61b954a91feeaad22601f43433abcf5fdc0a2cae694187a771609d1a73978458", "snapshot_at": "2025-03-15T17:36:14+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-MP-DocVQA/resolve/3ebd091c458cf04161f78cd7b12ea101f83e2529/qrels/docvqa_mp-eval-qrels.tsv"} +{"annotation_source": "VisRAG-Ret qrels derived from DocVQA-MP upstream annotations, Mathew et al. 2021", "bytes": 10533, "dataset": "openbmb/VisRAG-Ret-Test-MP-DocVQA", "lfs_pointer_bytes": 10533, "lfs_pointer_oid": "4c22754d456295207cc1e395742f603771e94e292c7c5f163b37c75c2c7e52e9", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "3ebd091c458cf04161f78cd7b12ea101f83e2529", "sha256": "4c22754d456295207cc1e395742f603771e94e292c7c5f163b37c75c2c7e52e9", "snapshot_at": "2025-03-15T17:36:14+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-MP-DocVQA/resolve/3ebd091c458cf04161f78cd7b12ea101f83e2529/qrels/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from DocVQA-MP upstream annotations, Mathew et al. 2021", "bytes": 40168, "dataset": "openbmb/VisRAG-Ret-Test-MP-DocVQA", "lfs_pointer_bytes": 40168, "lfs_pointer_oid": "571dd2d48ddfa3a3f097235914c5f0a411d2fdd91e3b411b61a9ebac017c92b3", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "queries/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "3ebd091c458cf04161f78cd7b12ea101f83e2529", "sha256": "571dd2d48ddfa3a3f097235914c5f0a411d2fdd91e3b411b61a9ebac017c92b3", "snapshot_at": "2025-03-15T17:36:14+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-MP-DocVQA/resolve/3ebd091c458cf04161f78cd7b12ea101f83e2529/queries/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from PlotQA upstream annotations, Matsuda et al. 2018", "bytes": 288987711, "dataset": "openbmb/VisRAG-Ret-Test-PlotQA", "lfs_pointer_bytes": 288987711, "lfs_pointer_oid": "d88ee0065017e740d67d586d26ba474c3b7142ac1643c8eb0573795acfbfd4d4", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "ef953ef8ab6d78ac112dd4cde6acdb2c2692039a", "sha256": "d88ee0065017e740d67d586d26ba474c3b7142ac1643c8eb0573795acfbfd4d4", "snapshot_at": "2025-03-15T17:38:38+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-PlotQA/resolve/ef953ef8ab6d78ac112dd4cde6acdb2c2692039a/corpus/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from PlotQA upstream annotations, Matsuda et al. 2018", "bytes": 274583, "dataset": "openbmb/VisRAG-Ret-Test-PlotQA", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/plotqa-eval-qrels.tsv", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "ef953ef8ab6d78ac112dd4cde6acdb2c2692039a", "sha256": "ba0e59f6893e1f964f57c01e13964001a144c16236d537b95a148d2b63d8d0cb", "snapshot_at": "2025-03-15T17:38:38+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-PlotQA/resolve/ef953ef8ab6d78ac112dd4cde6acdb2c2692039a/qrels/plotqa-eval-qrels.tsv"} +{"annotation_source": "VisRAG-Ret qrels derived from PlotQA upstream annotations, Matsuda et al. 2018", "bytes": 14786, "dataset": "openbmb/VisRAG-Ret-Test-PlotQA", "lfs_pointer_bytes": 14786, "lfs_pointer_oid": "2666420a48248b46169422ec4d89564db5f7789bd47251a04f488772c8bcaeb3", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "ef953ef8ab6d78ac112dd4cde6acdb2c2692039a", "sha256": "2666420a48248b46169422ec4d89564db5f7789bd47251a04f488772c8bcaeb3", "snapshot_at": "2025-03-15T17:38:38+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-PlotQA/resolve/ef953ef8ab6d78ac112dd4cde6acdb2c2692039a/qrels/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from PlotQA upstream annotations, Matsuda et al. 2018", "bytes": 49623, "dataset": "openbmb/VisRAG-Ret-Test-PlotQA", "lfs_pointer_bytes": 49623, "lfs_pointer_oid": "186a2017a1db7bebdf314d0c290919e51a179f8e06faff966061fc813098134f", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "queries/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:21+00:00", "revision": "ef953ef8ab6d78ac112dd4cde6acdb2c2692039a", "sha256": "186a2017a1db7bebdf314d0c290919e51a179f8e06faff966061fc813098134f", "snapshot_at": "2025-03-15T17:38:38+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-PlotQA/resolve/ef953ef8ab6d78ac112dd4cde6acdb2c2692039a/queries/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from SlideVQA upstream annotations, Yu et al. 2024 arXiv:2410.10594", "bytes": 104072676, "dataset": "openbmb/VisRAG-Ret-Test-SlideVQA", "lfs_pointer_bytes": 104072676, "lfs_pointer_oid": "d1a3275af3930103caa8b495f209c2bd35ad0c2e6b720afb8ad5430b476e9cc8", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "corpus/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "6b5f81a3f6953095934c26c5d420dfd039184d69", "sha256": "d1a3275af3930103caa8b495f209c2bd35ad0c2e6b720afb8ad5430b476e9cc8", "snapshot_at": "2025-03-15T17:40:55+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-SlideVQA/resolve/6b5f81a3f6953095934c26c5d420dfd039184d69/corpus/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from SlideVQA upstream annotations, Yu et al. 2024 arXiv:2410.10594", "bytes": 629671, "dataset": "openbmb/VisRAG-Ret-Test-SlideVQA", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/slidevqa-eval-qrels.tsv", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "6b5f81a3f6953095934c26c5d420dfd039184d69", "sha256": "ef52700548a9eee9bfcedef56a0009d2d9e2c28711fccff1ecf52cbb3d6dd8ef", "snapshot_at": "2025-03-15T17:40:55+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-SlideVQA/resolve/6b5f81a3f6953095934c26c5d420dfd039184d69/qrels/slidevqa-eval-qrels.tsv"} +{"annotation_source": "VisRAG-Ret qrels derived from SlideVQA upstream annotations, Yu et al. 2024 arXiv:2410.10594", "bytes": 47702, "dataset": "openbmb/VisRAG-Ret-Test-SlideVQA", "lfs_pointer_bytes": 47702, "lfs_pointer_oid": "abc676e9a797684e8cb002954c7536ee6fe05d5a8cc0378250d4653cf14e72d7", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "qrels/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "6b5f81a3f6953095934c26c5d420dfd039184d69", "sha256": "abc676e9a797684e8cb002954c7536ee6fe05d5a8cc0378250d4653cf14e72d7", "snapshot_at": "2025-03-15T17:40:55+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-SlideVQA/resolve/6b5f81a3f6953095934c26c5d420dfd039184d69/qrels/train-00000-of-00001.parquet"} +{"annotation_source": "VisRAG-Ret qrels derived from SlideVQA upstream annotations, Yu et al. 2024 arXiv:2410.10594", "bytes": 63259, "dataset": "openbmb/VisRAG-Ret-Test-SlideVQA", "lfs_pointer_bytes": 63259, "lfs_pointer_oid": "bcf3402abaf8b8139d2a0efc989b26cef17222c1fb3365eb01f7e8caead522e1", "license": "not-stated-in-card", "modality": "image", "newly_downloaded": false, "oid_match": true, "path": "queries/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "6b5f81a3f6953095934c26c5d420dfd039184d69", "sha256": "bcf3402abaf8b8139d2a0efc989b26cef17222c1fb3365eb01f7e8caead522e1", "snapshot_at": "2025-03-15T17:40:55+00:00", "url": "https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-SlideVQA/resolve/6b5f81a3f6953095934c26c5d420dfd039184d69/queries/train-00000-of-00001.parquet"} +{"annotation_source": "LongMemEval human-authored QA with evidence_session_ids and category labels, Wu et al. 2024 arXiv:2410.10813", "bytes": 277383467, "dataset": "xiaowu0162/longmemeval-cleaned", "lfs_pointer_bytes": 277383467, "lfs_pointer_oid": "d6f21ea9d60a0d56f34a05b609c79c88a451d2ae03597821ea3d5a9678c3a442", "license": "mit", "modality": "conversation", "newly_downloaded": false, "oid_match": true, "path": "longmemeval_s_cleaned.json", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "98d7416c24c778c2fee6e6f3006e7a073259d48f", "sha256": "d6f21ea9d60a0d56f34a05b609c79c88a451d2ae03597821ea3d5a9678c3a442", "snapshot_at": "2025-09-19T23:48:16+00:00", "url": "https://huggingface.co/datasets/xiaowu0162/longmemeval-cleaned/resolve/98d7416c24c778c2fee6e6f3006e7a073259d48f/longmemeval_s_cleaned.json"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 11542, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": ".gitattributes", "retrieved_at": "2026-10-06T15:26:11+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "35a65ebfd4a03a68defcacf6cdc786441256abb05b80a0fdece88e33a7d4c80c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/.gitattributes"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 511, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "README.md", "retrieved_at": "2026-10-06T15:26:11+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "56eabfe8193c1e7932dc1ed88c1128128473b8804593d3557ed4fce5f71d44c3", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/README.md"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 82073, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 82073, "lfs_pointer_oid": "bcdac3c96669634c34184814cede4fe57cf7ac0f98dde0e85936394f6a56a02d", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "data/train-00000-of-00001.parquet", "retrieved_at": "2026-10-06T15:26:11+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "bcdac3c96669634c34184814cede4fe57cf7ac0f98dde0e85936394f6a56a02d", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/data/train-00000-of-00001.parquet"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2531046, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2531046, "lfs_pointer_oid": "a1d9fc716060605412a10127c4e158febc9e2ae13af04cbcf88bcc789c474d49", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/05-03-18-political-release.pdf", "retrieved_at": "2026-10-06T15:26:11+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "a1d9fc716060605412a10127c4e158febc9e2ae13af04cbcf88bcc789c474d49", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/05-03-18-political-release.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 6789900, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 6789900, "lfs_pointer_oid": "004b6c63f504f3868b678e04339be94361434e405ff2cf54fbf84e894a71183c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/0b85477387a9d0cc33fca0f4becaa0e5.pdf", "retrieved_at": "2026-10-06T15:26:11+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "004b6c63f504f3868b678e04339be94361434e405ff2cf54fbf84e894a71183c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/0b85477387a9d0cc33fca0f4becaa0e5.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1336503, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1336503, "lfs_pointer_oid": "13ff22623b747d1facbb405369001c6d3793322b24a56365bb8cef89db2f903e", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/0e94b4197b10096b1f4c699701570fbf.pdf", "retrieved_at": "2026-10-06T15:26:11+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "13ff22623b747d1facbb405369001c6d3793322b24a56365bb8cef89db2f903e", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/0e94b4197b10096b1f4c699701570fbf.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5924004, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5924004, "lfs_pointer_oid": "8f49f93f1665fd5ff604e66ffed9649d8b18ce7f114ab43463aeac46cd3d00cb", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/11-21-16-Updated-Post-Election-Release.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "8f49f93f1665fd5ff604e66ffed9649d8b18ce7f114ab43463aeac46cd3d00cb", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/11-21-16-Updated-Post-Election-Release.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1652256, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1652256, "lfs_pointer_oid": "67a67fe7666cc3ffdf8517220eed96b1f6b180b25bea2f2761f9d957d460a619", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/12-15-15-ISIS-and-terrorism-release-final.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "67a67fe7666cc3ffdf8517220eed96b1f6b180b25bea2f2761f9d957d460a619", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/12-15-15-ISIS-and-terrorism-release-final.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 7634041, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 7634041, "lfs_pointer_oid": "d4b846bd44896e1d21527f48eddc422c263c9a4f8a3fa4be249750956882698d", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2005.12872v3.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d4b846bd44896e1d21527f48eddc422c263c9a4f8a3fa4be249750956882698d", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2005.12872v3.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5284032, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5284032, "lfs_pointer_oid": "6ec3f43d9efb3df34e3ee3f0f72c2a3497892bdaf960fc0e6ce6a507bbc08425", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2021-Apple-Catalog.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "6ec3f43d9efb3df34e3ee3f0f72c2a3497892bdaf960fc0e6ce6a507bbc08425", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2021-Apple-Catalog.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 712862, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2023.acl-long.386.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "0928396891b1b209c34dcc16f44f334beaec69cc18c8d15d4d411239d5e40e26", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2023.acl-long.386.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 768651, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2023.findings-emnlp.248.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "9c76e97f3009dac27c830f31c9f126f8650691919b711eab7658f4c1ea9c74b0", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2023.findings-emnlp.248.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3379283, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3379283, "lfs_pointer_oid": "c99d936c7f64f74a7d62b99120eae3c005e0f3dc69d67a07569cc1730570cadc", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2024.ug.eprospectus.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "c99d936c7f64f74a7d62b99120eae3c005e0f3dc69d67a07569cc1730570cadc", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2024.ug.eprospectus.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4697845, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4697845, "lfs_pointer_oid": "d21971c4900736af848def300a03954f54a5827c3d14dfef77b5abcd0819555d", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2210.02442v1.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d21971c4900736af848def300a03954f54a5827c3d14dfef77b5abcd0819555d", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2210.02442v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1731233, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1731233, "lfs_pointer_oid": "d12d658bde0b32460f392b15b75a5b3fa18b0d2528884909de1de34760cd84bc", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2303.05039v2.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d12d658bde0b32460f392b15b75a5b3fa18b0d2528884909de1de34760cd84bc", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2303.05039v2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1473753, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1473753, "lfs_pointer_oid": "14cbe24bdb2c54bc82811e9a0984780032ecbfd27a9b9871fb8e12921dcd159d", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2303.08559v2.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "14cbe24bdb2c54bc82811e9a0984780032ecbfd27a9b9871fb8e12921dcd159d", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2303.08559v2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2418832, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2418832, "lfs_pointer_oid": "84d0c4afccf1b776b7a3c752c9f4b18580d199ccdbadc67a27f1fc02e9ea77ac", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2305.13186v3.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "84d0c4afccf1b776b7a3c752c9f4b18580d199ccdbadc67a27f1fc02e9ea77ac", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2305.13186v3.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 6338813, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 6338813, "lfs_pointer_oid": "035b1659c5305a0c8157594cf41446fe72ade8d48efa99eb036dfa522d3431c4", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2305.14160v4.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "035b1659c5305a0c8157594cf41446fe72ade8d48efa99eb036dfa522d3431c4", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2305.14160v4.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 9805311, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 9805311, "lfs_pointer_oid": "570cc7bda7a340972d3d74f30de647039f3edca3fce42f992bca5ca7bf894579", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2306.05425v1.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "570cc7bda7a340972d3d74f30de647039f3edca3fce42f992bca5ca7bf894579", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2306.05425v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 13661300, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 13661300, "lfs_pointer_oid": "1df284ce95f783002074bfe8f21d47c646b396ceb1736ea3ec0ea212fc070d91", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2307.09288v2.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "1df284ce95f783002074bfe8f21d47c646b396ceb1736ea3ec0ea212fc070d91", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2307.09288v2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 45669728, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 45669728, "lfs_pointer_oid": "0789456681582c1bf7bbb3e03c47c022476d4bb723cfe840ed8a183fb67ef198", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2309.17421v2.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "0789456681582c1bf7bbb3e03c47c022476d4bb723cfe840ed8a183fb67ef198", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2309.17421v2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1847937, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1847937, "lfs_pointer_oid": "7cb9c5f18f753325319d7cacf5d5a53d39ce1cf368c28fc85455388955cf0dfe", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2310.05634v2.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7cb9c5f18f753325319d7cacf5d5a53d39ce1cf368c28fc85455388955cf0dfe", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2310.05634v2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1607888, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1607888, "lfs_pointer_oid": "d42888b4f18c3838dddf1d72e977fb500fea77354246b2f8d1c0a62f92b46bd2", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2310.07609v1.pdf", "retrieved_at": "2026-10-06T15:26:12+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d42888b4f18c3838dddf1d72e977fb500fea77354246b2f8d1c0a62f92b46bd2", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2310.07609v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2762223, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2762223, "lfs_pointer_oid": "45e8a907fb0c10450be68cd1711548df85f806d66a8bfea9eb7adf51f4a3e227", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2310.09158v1.pdf", "retrieved_at": "2026-10-06T15:26:13+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "45e8a907fb0c10450be68cd1711548df85f806d66a8bfea9eb7adf51f4a3e227", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2310.09158v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 16763496, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 16763496, "lfs_pointer_oid": "48d87afcb1770e273491f946193f339c3e86ae85b46064e6e322ec732d42f2d1", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2311.16502v3.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "48d87afcb1770e273491f946193f339c3e86ae85b46064e6e322ec732d42f2d1", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2311.16502v3.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1205850, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1205850, "lfs_pointer_oid": "6020ba90c4f16fb632cae4f698e565b25f713a58f13ebbdbd330ca5943b2fc24", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2312.04350v3.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "6020ba90c4f16fb632cae4f698e565b25f713a58f13ebbdbd330ca5943b2fc24", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2312.04350v3.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2299371, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2299371, "lfs_pointer_oid": "720656b49d4c00dac26d6ebe10dad0d8482d3bcc87640a66b73867730b89515f", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2312.09390v1.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "720656b49d4c00dac26d6ebe10dad0d8482d3bcc87640a66b73867730b89515f", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2312.09390v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1662567, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1662567, "lfs_pointer_oid": "396a0fadeb4cd40f5c8ccc36b73a0815f6cb4d7f6bfa53b6c48c1f9aba7c7e02", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2312.10997v5.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "396a0fadeb4cd40f5c8ccc36b73a0815f6cb4d7f6bfa53b6c48c1f9aba7c7e02", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2312.10997v5.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2547113, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2547113, "lfs_pointer_oid": "01319e2e6e6ecfd504ee1bbda41d023f9e060bd3505ab3c161baa1d6d696cf75", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2401.18059v1.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "01319e2e6e6ecfd504ee1bbda41d023f9e060bd3505ab3c161baa1d6d696cf75", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2401.18059v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 25242519, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 25242519, "lfs_pointer_oid": "60cc0caec7637ebd490e927f861500a9e191582f5babf1686d08ac454201ab6c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/2405.09818v1.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "60cc0caec7637ebd490e927f861500a9e191582f5babf1686d08ac454201ab6c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/2405.09818v1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1004229, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1004229, "lfs_pointer_oid": "03708bc596d91f8d0000bc44b0794b1300fffa24ab96419e361e483d0d47be93", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/3276a5b991c49cf5f9a4af0f7d6fce67.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "03708bc596d91f8d0000bc44b0794b1300fffa24ab96419e361e483d0d47be93", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/3276a5b991c49cf5f9a4af0f7d6fce67.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 287691, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/379f44022bb27aa53efd5d322c7b57bf.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "08408fea6869f71b8bb85b86143ded39b28fc8b32a9090ceb64211febd446a67", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/379f44022bb27aa53efd5d322c7b57bf.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1252586, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1252586, "lfs_pointer_oid": "faf10c53fc6f1a8aa346ec3db79145e3ae6fbb3753c8064924e174d0a973934d", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/3M_2018_10K.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "faf10c53fc6f1a8aa346ec3db79145e3ae6fbb3753c8064924e174d0a973934d", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/3M_2018_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 12263327, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 12263327, "lfs_pointer_oid": "b59a69ee9d61996ffb62e8c3db5f8ad41934c803f827f11d6b7da50f7e5026c4", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/52b3137455e7ca4df65021a200aef724.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "b59a69ee9d61996ffb62e8c3db5f8ad41934c803f827f11d6b7da50f7e5026c4", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/52b3137455e7ca4df65021a200aef724.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 455160, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/698bba535087fa9a7f9009e172a7f763.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "be8b8e31e4804cd3a8dc5e67f4cf877b0295ded30f9db250a5694c3264acbbe0", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/698bba535087fa9a7f9009e172a7f763.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 449018, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/7c3f6204b3241f142f0f8eb8e1fefe7a.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "c6cca153a7681b0af2b51f06f2528affe5b4198e3a3990a64cb0f62eb928f860", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/7c3f6204b3241f142f0f8eb8e1fefe7a.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2371555, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2371555, "lfs_pointer_oid": "b38932583846e5de2daadcf65376621de643541ad2ec54f37ec205e4ebfc8b80", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/8dfc21ec151fb9d3578fc32d5c4e5df9.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "b38932583846e5de2daadcf65376621de643541ad2ec54f37ec205e4ebfc8b80", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/8dfc21ec151fb9d3578fc32d5c4e5df9.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3094711, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3094711, "lfs_pointer_oid": "b5c5a7769d5fbcaf0ffc5bd7d67df17ce0c6c92dcadfa60cf34fff6d2c9161b0", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/8e7c4cb542ad160f80fb3d795ada35d8.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "b5c5a7769d5fbcaf0ffc5bd7d67df17ce0c6c92dcadfa60cf34fff6d2c9161b0", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/8e7c4cb542ad160f80fb3d795ada35d8.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1677202, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1677202, "lfs_pointer_oid": "a3dfbac95e5c99ea9aa9fb830bd5f338d3278c7df28a775bc314acedc22b20d5", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/91521110100M_4K_UHD_Display_User_Manual_V1.1.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "a3dfbac95e5c99ea9aa9fb830bd5f338d3278c7df28a775bc314acedc22b20d5", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/91521110100M_4K_UHD_Display_User_Manual_V1.1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 308854, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/936c0e2c2e6c8e0c07c51bfaf7fd0a83.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "c340ba25ac7d837244e99d7bd692ae79fd3280682287874eaf0748da60e03171", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/936c0e2c2e6c8e0c07c51bfaf7fd0a83.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1274941, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1274941, "lfs_pointer_oid": "017f1bbc1e37f2945a41fa2b5cf39d822e1e9d435a2c2b9c6b1cc76cd114a2df", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/ACTIVISIONBLIZZARD_2019_10K.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "017f1bbc1e37f2945a41fa2b5cf39d822e1e9d435a2c2b9c6b1cc76cd114a2df", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/ACTIVISIONBLIZZARD_2019_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1457725, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1457725, "lfs_pointer_oid": "f9ddba652f19e20752d98b1b4cedcebcb1a9a6d11dd23ab2e2f00f009a164c24", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/ADOBE_2015_10K.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "f9ddba652f19e20752d98b1b4cedcebcb1a9a6d11dd23ab2e2f00f009a164c24", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/ADOBE_2015_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 627047, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/AMAZON_2017_10K.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "f8af08ff6c4147351ec3a6e5e2c74e104678ce68fb0d6d476b1e43f685c09a3f", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/AMAZON_2017_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 703034, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/BESTBUY_2023_10K.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7ca465d31a08be29e8c08d756644d33d818c7254a1215720f265227005cda696", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/BESTBUY_2023_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4415868, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4415868, "lfs_pointer_oid": "5967410aa6e7e69f0b8dfbcf1dcd15e4ffda0a7f4b87b406c48d73755a19cfeb", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/BRO-GL-MMONEY.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "5967410aa6e7e69f0b8dfbcf1dcd15e4ffda0a7f4b87b406c48d73755a19cfeb", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/BRO-GL-MMONEY.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 11939314, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 11939314, "lfs_pointer_oid": "f1c950b0504468709292128095a0f7b503b8f4107a42e1258cda7bc01192de7d", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Bergen-Brochure-en-2022-23.pdf", "retrieved_at": "2026-10-06T15:26:14+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "f1c950b0504468709292128095a0f7b503b8f4107a42e1258cda7bc01192de7d", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Bergen-Brochure-en-2022-23.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 765319, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/COSTCO_2021_10K.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "29e05b3862cf6ffa6cfda34ac82a9fa0fe76b67338c5924da3325c2ee3b08748", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/COSTCO_2021_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3813435, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3813435, "lfs_pointer_oid": "185095809183fcc58ed8dee7eac4da1011e042639ffa1bf1bdd817ed5049e4a0", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Campaign_038_Introducing_AC_Whitepaper_v5e.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "185095809183fcc58ed8dee7eac4da1011e042639ffa1bf1bdd817ed5049e4a0", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Campaign_038_Introducing_AC_Whitepaper_v5e.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 771368, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/DSA-278777.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d7c46ad196d20c276142f7b7615beee9afc38b1f267c244440d1c0f76ee4d03b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/DSA-278777.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1256526, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1256526, "lfs_pointer_oid": "83424ce04cc38c567e121a1f6827e370a878917a8d8b5b7743e05f1c49042f5b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/GPL-Graduate-Studies-Professional-Learning-Brochure-Jul-2021.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "83424ce04cc38c567e121a1f6827e370a878917a8d8b5b7743e05f1c49042f5b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/GPL-Graduate-Studies-Professional-Learning-Brochure-Jul-2021.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 8266993, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 8266993, "lfs_pointer_oid": "19be7df5dce477f643b627e6ecb61f400acda0f9e7394a50b7a7a0730d0fbd2f", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Guide-for-international-students-web.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "19be7df5dce477f643b627e6ecb61f400acda0f9e7394a50b7a7a0730d0fbd2f", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Guide-for-international-students-web.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1432092, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1432092, "lfs_pointer_oid": "27163e80a71d12f267592d87c1112413ac85ed432c766f45e9558a47e6f9cb80", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/ISEP_student_handbook_2020.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "27163e80a71d12f267592d87c1112413ac85ed432c766f45e9558a47e6f9cb80", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/ISEP_student_handbook_2020.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 741327, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Independents-Report.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "bc1d3a07194f0423b4c2771bbb9bea45f80863518734fff7dab0930a6b467858", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Independents-Report.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3871985, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3871985, "lfs_pointer_oid": "d99de99131b4d8a5471d3e3d96befaa1e6152560d12a93f854c5dd95e54225aa", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Macbook_air.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d99de99131b4d8a5471d3e3d96befaa1e6152560d12a93f854c5dd95e54225aa", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Macbook_air.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 578489, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/NETFLIX_2015_10K.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "b22ca874ef08f383bd0b1071677d44f4a242af668f855f00dc9c6d4ac94148e2", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/NETFLIX_2015_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 877882, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/NIKE_2021_10K.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "73b54e5e0b1b0b33e51ac314f6d9a9db0bcc286269feb24a3445a5f1ff0f7789", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/NIKE_2021_10K.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3958495, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3958495, "lfs_pointer_oid": "482a559e9e2cae5a831c11565786ac274d67e0f5b1096927e7fdf1c25152358e", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/NUS-Business-School-BBA-Brochure-2024.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "482a559e9e2cae5a831c11565786ac274d67e0f5b1096927e7fdf1c25152358e", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/NUS-Business-School-BBA-Brochure-2024.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 15118768, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 15118768, "lfs_pointer_oid": "8e7802bfbb70e4126bcc6a5f4d06bcc1c549dd24cf4c90a143eab8971abe66a8", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/NUS-FASS-Graduate-Guidebook-2021-small.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "8e7802bfbb70e4126bcc6a5f4d06bcc1c549dd24cf4c90a143eab8971abe66a8", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/NUS-FASS-Graduate-Guidebook-2021-small.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 31841340, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 31841340, "lfs_pointer_oid": "16ee1cfbd58f9c6a3793c66ff5f4ec38dd864ac9fd7a9b7186f8af7db53f870b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/NYU_graduate.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "16ee1cfbd58f9c6a3793c66ff5f4ec38dd864ac9fd7a9b7186f8af7db53f870b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/NYU_graduate.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2620223, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2620223, "lfs_pointer_oid": "a529a14b12bfa3986b6b86c5ad26622f50e688332c03d4670ec9fe1b7076eb10", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PG_20.07.30_U.S.-Views-China_final.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "a529a14b12bfa3986b6b86c5ad26622f50e688332c03d4670ec9fe1b7076eb10", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PG_20.07.30_U.S.-Views-China_final.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 658530, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PG_2020.03.09_US-Germany_FINAL.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "538faebdf5ff106946523de5cf66c2a55dfd5f5cf1fa77558bf8bfad88ba4a86", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PG_2020.03.09_US-Germany_FINAL.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3297241, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3297241, "lfs_pointer_oid": "7b4ecd097f7fcb980c8cdcc1771915bcd0ff548de2de02194769d1c8c39d163c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PG_2020.05.21_International-Cooperation-COVID_FINAL.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7b4ecd097f7fcb980c8cdcc1771915bcd0ff548de2de02194769d1c8c39d163c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PG_2020.05.21_International-Cooperation-COVID_FINAL.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2734242, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2734242, "lfs_pointer_oid": "a1d2be648c1752c4ec86c398ae3e199fd0976a61c4fcd4060a5497750425b461", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PG_2021.03.04_US-Views-on-China_FINAL.pdf", "retrieved_at": "2026-10-06T15:26:15+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "a1d2be648c1752c4ec86c398ae3e199fd0976a61c4fcd4060a5497750425b461", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PG_2021.03.04_US-Views-on-China_FINAL.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2848610, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2848610, "lfs_pointer_oid": "10e57152f44d1a2b2d5325c4aafec4970b573b672ab253fb233731f327e7856c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PH_2016.06.08_Economy-Final.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "10e57152f44d1a2b2d5325c4aafec4970b573b672ab253fb233731f327e7856c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PH_2016.06.08_Economy-Final.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 525251, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PIP_Seniors-and-Tech-Use_040314.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "11fc8671985f4a810ed635ba2df7ae29c2fc0632de230b2a02386b2fbad08103", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PIP_Seniors-and-Tech-Use_040314.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 9370404, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 9370404, "lfs_pointer_oid": "9c60be28b53db3b25cb758c7c50285f1ce475f94f926b77e53f59633f26c589c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PI_2017.10.04_Automation_FINAL.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "9c60be28b53db3b25cb758c7c50285f1ce475f94f926b77e53f59633f26c589c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PI_2017.10.04_Automation_FINAL.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5395179, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5395179, "lfs_pointer_oid": "4f8f9048760c3ef36fd08737cdf925628b6401a38164655571ac7ac1ce5f2450", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PI_2018.11.19_algorithms_FINAL.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "4f8f9048760c3ef36fd08737cdf925628b6401a38164655571ac7ac1ce5f2450", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PI_2018.11.19_algorithms_FINAL.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4184545, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4184545, "lfs_pointer_oid": "35450aabe044f8af225c0ef26cb770a4e36e87623ed1185aed25853bc3ba90bb", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PP_2019.01.17_Trump-economy_FINAL2.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "35450aabe044f8af225c0ef26cb770a4e36e87623ed1185aed25853bc3ba90bb", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PP_2019.01.17_Trump-economy_FINAL2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3362691, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3362691, "lfs_pointer_oid": "db945aff713c1b81938db35a35cb27e3630c31c9f0a5c77d435df2d82e2d6ecf", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PP_2020.08.06_COVID-19-Restrictions_FINAL-1.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "db945aff713c1b81938db35a35cb27e3630c31c9f0a5c77d435df2d82e2d6ecf", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PP_2020.08.06_COVID-19-Restrictions_FINAL-1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3180371, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3180371, "lfs_pointer_oid": "192f585da0105856e0b56f286049fa5a23ae36b81bf54ae87ed2bb08b6f9778f", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PP_2021.04.22_voting-access_REPORT.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "192f585da0105856e0b56f286049fa5a23ae36b81bf54ae87ed2bb08b6f9778f", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PP_2021.04.22_voting-access_REPORT.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 13419451, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 13419451, "lfs_pointer_oid": "e04912074ec706c890bae376722b87b6cb626b2b53b33d43bc029d5a17ac7601", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PRE_2022.09.29_NSL-politics_REPORT.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "e04912074ec706c890bae376722b87b6cb626b2b53b33d43bc029d5a17ac7601", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PRE_2022.09.29_NSL-politics_REPORT.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 14287689, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 14287689, "lfs_pointer_oid": "6fff1e95a1b944a416a0b4e5d9d4112b9716106cfdd16c3b2b21d9bca8b6bafa", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PS_2018.01.09_STEM_FINAL.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "6fff1e95a1b944a416a0b4e5d9d4112b9716106cfdd16c3b2b21d9bca8b6bafa", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PS_2018.01.09_STEM_FINAL.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3751809, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3751809, "lfs_pointer_oid": "ba92752271eac6db634d84d9ae29840f08e252b71a7e2ab0dc068e9e91a4ba4b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/PWC_opportunity_of_lifetime.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "ba92752271eac6db634d84d9ae29840f08e252b71a7e2ab0dc068e9e91a4ba4b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/PWC_opportunity_of_lifetime.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4079034, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4079034, "lfs_pointer_oid": "29502cf5dbeeed56204d62786c9a6e8c55579d72bce57b84d370c9fff503d164", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Pew-Research-Center_Hispanic-Identity-Report_12.20.2017.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "29502cf5dbeeed56204d62786c9a6e8c55579d72bce57b84d370c9fff503d164", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Pew-Research-Center_Hispanic-Identity-Report_12.20.2017.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5453220, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5453220, "lfs_pointer_oid": "154b7ddf558e01be8b11ded9aec5645b5a8db1052877e2c0aa6909cf9596377c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/RAR.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "154b7ddf558e01be8b11ded9aec5645b5a8db1052877e2c0aa6909cf9596377c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/RAR.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 8813581, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 8813581, "lfs_pointer_oid": "7515cd264f0a33a782169d938a291e9ec4fc81a094a30a4ecf951d4cfa7ff336", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/SAO-StudentSupport_Guidebook-Content.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7515cd264f0a33a782169d938a291e9ec4fc81a094a30a4ecf951d4cfa7ff336", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/SAO-StudentSupport_Guidebook-Content.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1024673, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1024673, "lfs_pointer_oid": "1068e568b292ce57a6e980eab659fd084eb8a3c03bcedb2164cafc35337b5ad2", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/STEPBACK.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "1068e568b292ce57a6e980eab659fd084eb8a3c03bcedb2164cafc35337b5ad2", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/STEPBACK.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 19410173, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 19410173, "lfs_pointer_oid": "850436eff14422e68332c66ad72c58232afe8983ce15dfb04a848a742ca4f0db", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/Sinopolis-Chengdu.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "850436eff14422e68332c66ad72c58232afe8983ce15dfb04a848a742ca4f0db", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/Sinopolis-Chengdu.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 17391142, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 17391142, "lfs_pointer_oid": "595241adee0bb0612dc89c287240b8c8b59d1c7f15fe6c15b04e7e43d722a1b9", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/SnapNTell.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "595241adee0bb0612dc89c287240b8c8b59d1c7f15fe6c15b04e7e43d722a1b9", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/SnapNTell.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 8813581, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 8813581, "lfs_pointer_oid": "7515cd264f0a33a782169d938a291e9ec4fc81a094a30a4ecf951d4cfa7ff336", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/StudentSupport_Guidebook.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7515cd264f0a33a782169d938a291e9ec4fc81a094a30a4ecf951d4cfa7ff336", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/StudentSupport_Guidebook.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1789965, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1789965, "lfs_pointer_oid": "ba6a07e262b070f963cf1f4ef3372074d08e5820a833f130e67d597af211359c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/User_Manual_1500S_Classic_EN.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "ba6a07e262b070f963cf1f4ef3372074d08e5820a833f130e67d597af211359c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/User_Manual_1500S_Classic_EN.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 100113, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/a4f3ced0696009fec3179f493e4f28c4.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "702351312f3241cc51d140064ba02ac63ec50581c3a8a664c9249908d8c0c30b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/a4f3ced0696009fec3179f493e4f28c4.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 377310, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/a5879805d70c854ea4361e43a84e3bb2.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "b49aac986f8b13cdca3c429013ab668304e64eca8c72e85f9961acf3616d793e", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/a5879805d70c854ea4361e43a84e3bb2.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 522738, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/afe620b9beac86c1027b96d31d396407.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "a7b0ae685a743a0acedd57477fd20e3ad0a09e15ec73203b83ea531facb4e5d9", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/afe620b9beac86c1027b96d31d396407.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2929623, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2929623, "lfs_pointer_oid": "181c9f34668397c8ff0e5e0ca7ceed59f2ec023a7fb51ee204983a57e2af779c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/amb-siteaudits-ds15-150204174043-conversion-gate01_95.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "181c9f34668397c8ff0e5e0ca7ceed59f2ec023a7fb51ee204983a57e2af779c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/amb-siteaudits-ds15-150204174043-conversion-gate01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4472720, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4472720, "lfs_pointer_oid": "737e16a1aaac8ba108e7dc92d542960b93c07fe723593920b528557394e4986b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/asdaaburson-marstellerarabyouthsurvey2014-140407100615-phpapp01_95.pdf", "retrieved_at": "2026-10-06T15:26:16+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "737e16a1aaac8ba108e7dc92d542960b93c07fe723593920b528557394e4986b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/asdaaburson-marstellerarabyouthsurvey2014-140407100615-phpapp01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1960940, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1960940, "lfs_pointer_oid": "40e8219add5c83ebe97ce47a3a8e4bce3dd1c27ecc592f897dba0f3a4a6ec831", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/avalaunchpresentationsthatkickasteriskv3copy-150318114804-conversion-gate01_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "40e8219add5c83ebe97ce47a3a8e4bce3dd1c27ecc592f897dba0f3a4a6ec831", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/avalaunchpresentationsthatkickasteriskv3copy-150318114804-conversion-gate01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1906963, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1906963, "lfs_pointer_oid": "43be9592854c7da10bf011335faec017b726ff2e23bb4ccae7e8fe88d7801f22", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/b3m5kaeqm2w8n4bwcesw-140602121350-phpapp02_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "43be9592854c7da10bf011335faec017b726ff2e23bb4ccae7e8fe88d7801f22", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/b3m5kaeqm2w8n4bwcesw-140602121350-phpapp02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 7498433, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 7498433, "lfs_pointer_oid": "8fc1f4121b4fa69f2f87b37a6236bc40578b0a93e5c94fa886cfef751e0f11b6", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/bariumswallowpresentation-090810084400-phpapp01_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "8fc1f4121b4fa69f2f87b37a6236bc40578b0a93e5c94fa886cfef751e0f11b6", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/bariumswallowpresentation-090810084400-phpapp01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 7669178, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 7669178, "lfs_pointer_oid": "d012fa4f6b35eab5507789709b08101d05c0993cc43ac9495165ef896620018b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/bdf54dxa.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d012fa4f6b35eab5507789709b08101d05c0993cc43ac9495165ef896620018b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/bdf54dxa.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2891511, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2891511, "lfs_pointer_oid": "8cbcce574c75ffb1e78905910616493ad88242e98047aa93ec648a798a64ece3", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/bigdatatrends-120723191058-phpapp02_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "8cbcce574c75ffb1e78905910616493ad88242e98047aa93ec648a798a64ece3", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/bigdatatrends-120723191058-phpapp02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2814575, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2814575, "lfs_pointer_oid": "972056ce39de67fc3eef402099a8f58fa3388a5142b4314a286968e7999bbc74", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/c31e6580d0175ab3f9d99d1ff0bfa000.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "972056ce39de67fc3eef402099a8f58fa3388a5142b4314a286968e7999bbc74", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/c31e6580d0175ab3f9d99d1ff0bfa000.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2238680, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2238680, "lfs_pointer_oid": "ebe07324dff37acdb4f322bcea916736ce911a4b9e0678099a4e4a0ce2e0f418", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/caltraincapacitymountainview1-150701205750-lva1-app6891_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "ebe07324dff37acdb4f322bcea916736ce911a4b9e0678099a4e4a0ce2e0f418", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/caltraincapacitymountainview1-150701205750-lva1-app6891_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3601732, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3601732, "lfs_pointer_oid": "d3f8049bc0c48b735b7c4e0d6399ed9e8188afd332f58829f44e1c39cfb13528", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/camry_ebrochure.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d3f8049bc0c48b735b7c4e0d6399ed9e8188afd332f58829f44e1c39cfb13528", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/camry_ebrochure.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5296218, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5296218, "lfs_pointer_oid": "40a74d3fc5e73c81465206a4043c19b850daf66fc5617756f5894d25ad054b4b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/catvsdogdlpycon15se-150512122612-lva1-app6891_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "40a74d3fc5e73c81465206a4043c19b850daf66fc5617756f5894d25ad054b4b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/catvsdogdlpycon15se-150512122612-lva1-app6891_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5549617, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5549617, "lfs_pointer_oid": "c4955415e7229ad3460f572dc7a4eac9c82ccdab0156cf22069f52046e5cb914", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/chapter8-geneticscompatibilitymode-141214140247-conversion-gate02_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "c4955415e7229ad3460f572dc7a4eac9c82ccdab0156cf22069f52046e5cb914", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/chapter8-geneticscompatibilitymode-141214140247-conversion-gate02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 5231491, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 5231491, "lfs_pointer_oid": "964fc4748a7698dd5af398a714f97cf430ff93f8646b75021c43bfb392fae870", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/competitiveoutcomes-091006065143-phpapp01_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "964fc4748a7698dd5af398a714f97cf430ff93f8646b75021c43bfb392fae870", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/competitiveoutcomes-091006065143-phpapp01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3658047, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3658047, "lfs_pointer_oid": "68edff324f440da6f72a5372bc622f0f81ae3a243b921c6798912402368303df", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/csewt7zsecmmbzjufbyx-signature-24d91a254426c21c3079384270e1f138dc43a271cfe15d6d520d68205855b2a3-poli-150306115347-conversion-gate01_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "68edff324f440da6f72a5372bc622f0f81ae3a243b921c6798912402368303df", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/csewt7zsecmmbzjufbyx-signature-24d91a254426c21c3079384270e1f138dc43a271cfe15d6d520d68205855b2a3-poli-150306115347-conversion-gate01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2948693, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2948693, "lfs_pointer_oid": "2495123a46df51c4c552350d7bdaefc5a3f0dfb5f16d5b741b65b9bd7126f547", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/ddoseattle-150627210357-lva1-app6891_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "2495123a46df51c4c552350d7bdaefc5a3f0dfb5f16d5b741b65b9bd7126f547", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/ddoseattle-150627210357-lva1-app6891_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 15398470, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 15398470, "lfs_pointer_oid": "13562992665a0bcb5b7ae43f633c32005b03a78f189db00890cefdd4db862ba8", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/digitalmeasurementframework22feb2011v6novideo-110221233835-phpapp01_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "13562992665a0bcb5b7ae43f633c32005b03a78f189db00890cefdd4db862ba8", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/digitalmeasurementframework22feb2011v6novideo-110221233835-phpapp01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2712146, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2712146, "lfs_pointer_oid": "1d2c0cb82cb55ae00a8121bc6fa476cb16057bd7c291cdd34323852ba125736c", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/disciplined-agile-business-analysis-160218012713_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "1d2c0cb82cb55ae00a8121bc6fa476cb16057bd7c291cdd34323852ba125736c", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/disciplined-agile-business-analysis-160218012713_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 15398470, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 11191551, "lfs_pointer_oid": "13562992665a0bcb5b7ae43f633c32005b03a78f189db00890cefdd4db862ba8", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/dr-vorapptchapter1emissionsources-121120210508-phpapp02_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "13562992665a0bcb5b7ae43f633c32005b03a78f189db00890cefdd4db862ba8", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/dr-vorapptchapter1emissionsources-121120210508-phpapp02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 774511, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/e639029d16094ea71d964e2fb953952b.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "773e1fc44166300d1473853e48b2b0c3f16164e3e125b52db8c616a34fdf8ea7", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/e639029d16094ea71d964e2fb953952b.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 475553, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/e79deb02a0c0e87511080836c5d4347b.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "ca33492fafca0831b1a248f5ce5e8015cbe1f858100e8ebce55c7224f32a175e", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/e79deb02a0c0e87511080836c5d4347b.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4037538, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4037538, "lfs_pointer_oid": "5d2e813c9f5ff982ed03b16f925fa9b493e788941dbeb2569d83d82e40f9d904", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/earlybird-110722143746-phpapp02_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "5d2e813c9f5ff982ed03b16f925fa9b493e788941dbeb2569d83d82e40f9d904", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/earlybird-110722143746-phpapp02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2943084, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2943084, "lfs_pointer_oid": "9548d570d5a26812882f7e4a9d022b00dd993fe9f11479d540a189b85ac93d75", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/earthlinkweb-150213112111-conversion-gate02_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "9548d570d5a26812882f7e4a9d022b00dd993fe9f11479d540a189b85ac93d75", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/earthlinkweb-150213112111-conversion-gate02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1790232, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1790232, "lfs_pointer_oid": "711e018acdc5c26fb0847bf98296d4f3935080d72bc959d7104a183ee376618f", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/ecommerceopportunityindia-141124010546-conversion-gate01_95.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "711e018acdc5c26fb0847bf98296d4f3935080d72bc959d7104a183ee376618f", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/ecommerceopportunityindia-141124010546-conversion-gate01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 9088735, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 9088735, "lfs_pointer_oid": "c7ff0cf6af8e08ec6075f7bae1c4272e170058ac308bb43faa64194a38c6369f", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/edb88a99670417f64a6b719646aed326.pdf", "retrieved_at": "2026-10-06T15:26:17+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "c7ff0cf6af8e08ec6075f7bae1c4272e170058ac308bb43faa64194a38c6369f", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/edb88a99670417f64a6b719646aed326.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1291838, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1291838, "lfs_pointer_oid": "6c5b29bf0cba703de3ce7baa19aaf61771596e906d388cee5fb40a1bda740d18", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/efd88e41c5f2606c57929cac6c1c0605.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "6c5b29bf0cba703de3ce7baa19aaf61771596e906d388cee5fb40a1bda740d18", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/efd88e41c5f2606c57929cac6c1c0605.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2783434, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2783434, "lfs_pointer_oid": "fedeac3e243be354fc875321e5f168441f2032e88ccadb9875267cb7cf7d67d6", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/efis-140411041451-phpapp01_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "fedeac3e243be354fc875321e5f168441f2032e88ccadb9875267cb7cf7d67d6", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/efis-140411041451-phpapp01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3660938, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3660938, "lfs_pointer_oid": "120054270633261077812b834bf62fe3d60d812adcd750c9f082e71a12725622", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/f1f5242528411b262be447e61e2eb10f.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "120054270633261077812b834bf62fe3d60d812adcd750c9f082e71a12725622", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/f1f5242528411b262be447e61e2eb10f.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 272719, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/f86d073b0d735ac873a65d906ba82758.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "712e7090d0c3e1001c184ce23f7c371e3b706a0d1bda4e6de09147160875e0e4", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/f86d073b0d735ac873a65d906ba82758.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 221641, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/f8d3a162ab9507e021d83dd109118b60.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "f2eb17a3ad57b7cf547da596a67f71b00c00d4c7e93d03a5ad9ec2d16c5618dd", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/f8d3a162ab9507e021d83dd109118b60.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1586692, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1586692, "lfs_pointer_oid": "3fc1c32fe534bc49d025c8699be4421a2abba60c71bdc360121c1f974e36a819", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/fd76bbefe469561966e5387aa709c482.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "3fc1c32fe534bc49d025c8699be4421a2abba60c71bdc360121c1f974e36a819", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/fd76bbefe469561966e5387aa709c482.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1295962, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1295962, "lfs_pointer_oid": "df65695cae4e30afe465edf187d334d55fa9a2c0ac7d72269cef09b38d2d39f1", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/fdac8d1e9ef56519371df7e6532df27d.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "df65695cae4e30afe465edf187d334d55fa9a2c0ac7d72269cef09b38d2d39f1", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/fdac8d1e9ef56519371df7e6532df27d.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3021284, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3021284, "lfs_pointer_oid": "e63ef520ce513cb19132f57d6efe8a59cf180a5a2aee4bbff1c4718804b06caa", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/finalmediafindingspdf-141228031149-conversion-gate02_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "e63ef520ce513cb19132f57d6efe8a59cf180a5a2aee4bbff1c4718804b06caa", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/finalmediafindingspdf-141228031149-conversion-gate02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1702874, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1702874, "lfs_pointer_oid": "68e5cabd0df9afe788abadb337e5666e9dbfd458a777d25df370d3ff4a52a863", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/finalpresentationdeck-whatwhyhowofcertificationsocial-160324220748_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "68e5cabd0df9afe788abadb337e5666e9dbfd458a777d25df370d3ff4a52a863", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/finalpresentationdeck-whatwhyhowofcertificationsocial-160324220748_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4985671, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4985671, "lfs_pointer_oid": "7eefebebdb9c5b0f6ba96e7e0e493123e58955ad60f380e0f2bbe3c7427ab366", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/formwork-150318073913-conversion-gate01_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7eefebebdb9c5b0f6ba96e7e0e493123e58955ad60f380e0f2bbe3c7427ab366", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/formwork-150318073913-conversion-gate01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1464468, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1464468, "lfs_pointer_oid": "9304adb2d85ad6065e05f832a6f315ea4f74bb4c30080fd103a0a0e014ae7111", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/germanwingsdigitalcrisisanalysis-150403064828-conversion-gate01_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "9304adb2d85ad6065e05f832a6f315ea4f74bb4c30080fd103a0a0e014ae7111", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/germanwingsdigitalcrisisanalysis-150403064828-conversion-gate01_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 19651858, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 19651858, "lfs_pointer_oid": "b82471e5adf0c53bfeebddedf382cf81ee4ee2e3d8f4b21829adf525e7374463", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/guojixueshengshenghuozhinanyingwen9.1.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "b82471e5adf0c53bfeebddedf382cf81ee4ee2e3d8f4b21829adf525e7374463", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/guojixueshengshenghuozhinanyingwen9.1.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 724735, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/honor_watch_gs_pro.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "2d5559e514dc54f616ffea8d8e054117e32da672bdbf9797883dd815fa990ee5", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/honor_watch_gs_pro.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 3342422, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 3342422, "lfs_pointer_oid": "a0635f60d506ec1acaed5f331de89dec23e47c2184fd8f52e802def3540a81be", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/indonesiamobilemarketresearch-ag-150106055934-conversion-gate02_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "a0635f60d506ec1acaed5f331de89dec23e47c2184fd8f52e802def3540a81be", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/indonesiamobilemarketresearch-ag-150106055934-conversion-gate02_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1610338, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1610338, "lfs_pointer_oid": "8d5e3496c76816158c84abc2155284bae16f713c7501bf8da10e58a2bf871374", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/measuringsuccessonfacebooktwitterlinkedin-160317142140_95.pdf", "retrieved_at": "2026-10-06T15:26:18+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "8d5e3496c76816158c84abc2155284bae16f713c7501bf8da10e58a2bf871374", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/measuringsuccessonfacebooktwitterlinkedin-160317142140_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 31841340, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 24877442, "lfs_pointer_oid": "c5feec13dc1033920affd4fb6dff558d265f917fd19aaa21fe39ce15b31a1712", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": false, "path": "documents/mi_phone.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "16ee1cfbd58f9c6a3793c66ff5f4ec38dd864ac9fd7a9b7186f8af7db53f870b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/mi_phone.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1539783, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1539783, "lfs_pointer_oid": "8d7fccbf5690ad72fadcffda6f4a7aa808a79853398addf89f765e5f34ab8d9b", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/mmdetection-readthedocs-io-en-v2.18.0.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "8d7fccbf5690ad72fadcffda6f4a7aa808a79853398addf89f765e5f34ab8d9b", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/mmdetection-readthedocs-io-en-v2.18.0.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2805732, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2805732, "lfs_pointer_oid": "10adef9ab39eba5eb3679669e323145147d121892b1d3efd65407be301b40c97", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/nielsen2015musicbizpresentation-final-150526143534-lva1-app6891_95.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "10adef9ab39eba5eb3679669e323145147d121892b1d3efd65407be301b40c97", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/nielsen2015musicbizpresentation-final-150526143534-lva1-app6891_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4112562, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4112562, "lfs_pointer_oid": "1d645b8b4b035ca54e197309d3d0db23e89cfc406066f030db44ee2cd0e536a5", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/nova_y70.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "1d645b8b4b035ca54e197309d3d0db23e89cfc406066f030db44ee2cd0e536a5", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/nova_y70.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1146273, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1146273, "lfs_pointer_oid": "67356c31d3733fc786b769a585ee7a2e8aae4e98b0dc113e8511d45832c099e9", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/obs-productdesc-en.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "67356c31d3733fc786b769a585ee7a2e8aae4e98b0dc113e8511d45832c099e9", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/obs-productdesc-en.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1572727, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1572727, "lfs_pointer_oid": "91e70aaf49192b4deb640154a50d0b14976c831be11a7f858f9e953a761b8681", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/owners-manual-2170416.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "91e70aaf49192b4deb640154a50d0b14976c831be11a7f858f9e953a761b8681", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/owners-manual-2170416.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 4266409, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 4266409, "lfs_pointer_oid": "6371f5760d51eef3ff66d6c4217ed8d3613e30cc59800c262511bc641db8d8c5", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/q1-2023-bilibili-inc-investor-presentation.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "6371f5760d51eef3ff66d6c4217ed8d3613e30cc59800c262511bc641db8d8c5", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/q1-2023-bilibili-inc-investor-presentation.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1618840, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1618840, "lfs_pointer_oid": "2c9691d3a9dcd6bf400d3b20d2fc7a24bdf01bf0195b6e1ac602997847e2d021", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/reportq32015-151009093138-lva1-app6891_95.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "2c9691d3a9dcd6bf400d3b20d2fc7a24bdf01bf0195b6e1ac602997847e2d021", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/reportq32015-151009093138-lva1-app6891_95.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2028300, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2028300, "lfs_pointer_oid": "e538e1e713684742ce5d6686efefc8e5c182e6f62f1a79f59ed0b2f0bc1e7e80", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/san-francisco-11-contents.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "e538e1e713684742ce5d6686efefc8e5c182e6f62f1a79f59ed0b2f0bc1e7e80", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/san-francisco-11-contents.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 2730573, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 2730573, "lfs_pointer_oid": "d17cde8229847cb53926e578d70410cf46f9dbfb8a0d62cf675e18c3625507e3", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/stereo_headset.pdf", "retrieved_at": "2026-10-06T15:26:19+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "d17cde8229847cb53926e578d70410cf46f9dbfb8a0d62cf675e18c3625507e3", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/stereo_headset.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 8817249, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 8817249, "lfs_pointer_oid": "ee0393dadb951a955032b56dc4ac1a0b34a1ad425814b9f301c1b8f483b5f675", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/t480_ug_en.pdf", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "ee0393dadb951a955032b56dc4ac1a0b34a1ad425814b9f301c1b8f483b5f675", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/t480_ug_en.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 1109602, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 1109602, "lfs_pointer_oid": "40b3ef43cd3a003e5f9ed00426ee681d6d83cfa9c4a175f676d80b8fafd8c0cf", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/tacl_a_00660.pdf", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "40b3ef43cd3a003e5f9ed00426ee681d6d83cfa9c4a175f676d80b8fafd8c0cf", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/tacl_a_00660.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 8338928, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 8338928, "lfs_pointer_oid": "659e609bbae0bd631b18a1f20e58cbcaab083a4e99b5475797e95958308cdb85", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/transform-software-delivery-with-valueedge-brochure.pdf", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "659e609bbae0bd631b18a1f20e58cbcaab083a4e99b5475797e95958308cdb85", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/transform-software-delivery-with-valueedge-brochure.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 389268, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": null, "lfs_pointer_oid": null, "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/watch_d.pdf", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "bb5fd3576ac080c867f200ddc90311cdb351fb8240c70c9468dc433936c136e6", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/watch_d.pdf"} +{"annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", "bytes": 8048776, "dataset": "yubo2333/MMLongBench-Doc", "lfs_pointer_bytes": 8048776, "lfs_pointer_oid": "7dea5266e6c8eaa3e3549c6796dfab684cc7fa808e14386d6cf1e413479a61b3", "license": "not-stated-in-card", "modality": "pdf", "newly_downloaded": false, "oid_match": true, "path": "documents/welcome-to-nus.pdf", "retrieved_at": "2026-10-06T15:26:20+00:00", "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", "sha256": "7dea5266e6c8eaa3e3549c6796dfab684cc7fa808e14386d6cf1e413479a61b3", "snapshot_at": "2025-11-06T05:23:19+00:00", "url": "https://huggingface.co/datasets/yubo2333/MMLongBench-Doc/resolve/2ff6aa9237fc777b6627dc57a486e9225ac5fb86/documents/welcome-to-nus.pdf"} diff --git a/bench/corpus_sources.py b/bench/corpus_sources.py new file mode 100644 index 0000000..1ff671d --- /dev/null +++ b/bench/corpus_sources.py @@ -0,0 +1,184 @@ +"""Pinned corpus sources for the ely-eye benchmark.""" + +from typing import Any + +SOURCES: list[dict[str, Any]] = [ + { + "dataset": "yubo2333/MMLongBench-Doc", + "revision": "2ff6aa9237fc777b6627dc57a486e9225ac5fb86", + "snapshot_at": "2025-11-06T05:23:19+00:00", + "modality": "pdf", + "license": "not-stated-in-card", + "annotation_source": "MMLongBench-Doc human-written questions with evidence_page_ids, Wang et al. 2024 arXiv:2407.01523", + "files": "ALL", + }, + { + "dataset": "xiaowu0162/longmemeval-cleaned", + "revision": "98d7416c24c778c2fee6e6f3006e7a073259d48f", + "snapshot_at": "2025-09-19T23:48:16+00:00", + "modality": "conversation", + "license": "mit", + "annotation_source": "LongMemEval human-authored QA with evidence_session_ids and category labels, Wu et al. 2024 arXiv:2410.10813", + "files": ["longmemeval_s_cleaned.json"], + }, + { + "dataset": "openbmb/VisRAG-Ret-Test-SlideVQA", + "revision": "6b5f81a3f6953095934c26c5d420dfd039184d69", + "snapshot_at": "2025-03-15T17:40:55+00:00", + "modality": "image", + "license": "not-stated-in-card", + "annotation_source": "VisRAG-Ret qrels derived from SlideVQA upstream annotations, Yu et al. 2024 arXiv:2410.10594", + "files": [ + "corpus/train-00000-of-00001.parquet", + "queries/train-00000-of-00001.parquet", + "qrels/train-00000-of-00001.parquet", + "qrels/slidevqa-eval-qrels.tsv", + ], + }, + { + "dataset": "openbmb/VisRAG-Ret-Test-InfoVQA", + "revision": "04ee2db1d0a05304899453a3b03baef5e18991a3", + "snapshot_at": "2025-03-15T17:37:21+00:00", + "modality": "image", + "license": "not-stated-in-card", + "annotation_source": "VisRAG-Ret qrels derived from InfoVQA upstream annotations, Mathew et al. 2021", + "files": [ + "corpus/train-00000-of-00001.parquet", + "queries/train-00000-of-00001.parquet", + "qrels/train-00000-of-00001.parquet", + ], + }, + { + "dataset": "openbmb/VisRAG-Ret-Test-ChartQA", + "revision": "31f5ceb5d60b02e065bff394cb582f5bbb01a9b6", + "snapshot_at": "2025-03-15T17:58:47+00:00", + "modality": "image", + "license": "not-stated-in-card", + "annotation_source": "VisRAG-Ret qrels derived from ChartQA upstream annotations, Masry et al. 2022", + "files": [ + "corpus/train-00000-of-00001.parquet", + "queries/train-00000-of-00001.parquet", + "qrels/train-00000-of-00001.parquet", + ], + }, + { + "dataset": "openbmb/VisRAG-Ret-Test-MP-DocVQA", + "revision": "3ebd091c458cf04161f78cd7b12ea101f83e2529", + "snapshot_at": "2025-03-15T17:36:14+00:00", + "modality": "image", + "license": "not-stated-in-card", + "annotation_source": "VisRAG-Ret qrels derived from DocVQA-MP upstream annotations, Mathew et al. 2021", + "files": [ + "corpus/train-00000-of-00001.parquet", + "queries/train-00000-of-00001.parquet", + "qrels/train-00000-of-00001.parquet", + "qrels/docvqa_mp-eval-qrels.tsv", + ], + }, + { + "dataset": "openbmb/VisRAG-Ret-Test-PlotQA", + "revision": "ef953ef8ab6d78ac112dd4cde6acdb2c2692039a", + "snapshot_at": "2025-03-15T17:38:38+00:00", + "modality": "image", + "license": "not-stated-in-card", + "annotation_source": "VisRAG-Ret qrels derived from PlotQA upstream annotations, Matsuda et al. 2018", + "files": [ + "corpus/train-00000-of-00001.parquet", + "queries/train-00000-of-00001.parquet", + "qrels/train-00000-of-00001.parquet", + "qrels/plotqa-eval-qrels.tsv", + ], + }, + { + "dataset": "openbmb/VisRAG-Ret-Test-ArxivQA", + "revision": "00a31fc1e899b5f81d7760eca4907b1a1005865b", + "snapshot_at": "2025-03-15T17:31:12+00:00", + "modality": "image", + "license": "not-stated-in-card", + "annotation_source": "VisRAG-Ret qrels derived from ArXivQA upstream annotations, Li et al. 2024", + "files": [ + "corpus/train-00000-of-00003.parquet", + "corpus/train-00001-of-00003.parquet", + "corpus/train-00002-of-00003.parquet", + "queries/train-00000-of-00001.parquet", + "qrels/train-00000-of-00001.parquet", + "qrels/arxivqa-eval-qrels.tsv", + ], + }, + { + "dataset": "code-search-net/code_search_net", + "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555", + "snapshot_at": "2026-02-23T14:11:41+00:00", + "modality": "code", + "license": "other (per-repository OSS licenses recorded in the dataset's metadata column)", + "annotation_source": "developer-written docstrings, Husain et al. 2019 (arXiv:1909.09436)", + "files": [ + "python/test-00000-of-00001.parquet", + "python/validation-00000-of-00001.parquet", + ], + }, + { + "dataset": "OS-Copilot/ScreenSpot-v2", + "revision": "5efbb1f1b5463a575f2eb7bc30fe29e49c15f93c", + "snapshot_at": "2025-02-10T11:15:38+00:00", + "modality": "ui_screenshot", + "license": "apache-2.0", + "annotation_source": "ScreenSpot human-written referring instructions with element bounding boxes, Li et al. 2023 arXiv:2308.08155", + "files": [ + "screenspotv2_image.zip", + "screenspot_desktop_v2.json", + "screenspot_web_v2.json", + "screenspot_mobile_v2.json", + ], + }, + { + "dataset": "lmms-eval/Video-MME", + "revision": "ead1408f75b618502df9a1d8e0950166bf0a2a0b", + "snapshot_at": "2024-07-04T08:14:20+00:00", + "modality": "video", + "license": "not-stated-in-card", + "annotation_source": "Video-MME human-written video QA with duration tiers and subtitle timings, Fu et al. 2024 arXiv:2405.21083", + "files": [ + "videomme/test-00000-of-00001.parquet", + "subtitle.zip", + "videos_chunked_01.zip", + "videos_chunked_02.zip", + "videos_chunked_03.zip", + ], + }, +] + +def source_for(dataset: str) -> dict[str, Any]: + """The pinned entry for a dataset: revision, snapshot date, licence, provenance.""" + for source in SOURCES: + if source["dataset"] == dataset: + return source + raise SystemExit(f"{dataset} is not a pinned corpus source") + + +GIT_REPOS = [ + { + "repo": "https://github.com/pyparsing/pyparsing", + "ref": "master", + "sha": "f803ae834288032a700331977639fc49c62fcd7f", + "language": "python", + "modality": "code", + "annotation_source": "symbol definitions parsed from upstream source, no model judgement", + }, + { + "repo": "https://github.com/rust-lang/regex", + "ref": "master", + "sha": "72d650cb0a880a01ab6dc2137c0888e8f89740f7", + "language": "rust", + "modality": "code", + "annotation_source": "symbol definitions parsed from upstream source, no model judgement", + }, + { + "repo": "https://github.com/tokio-rs/tokio", + "ref": "master", + "sha": "b2636752450484955e7ad334bac678424d51bc4a", + "language": "rust", + "modality": "code", + "annotation_source": "symbol definitions parsed from upstream source, no model judgement", + }, +] diff --git a/bench/fetch_corpus.py b/bench/fetch_corpus.py new file mode 100644 index 0000000..0ac702c --- /dev/null +++ b/bench/fetch_corpus.py @@ -0,0 +1,267 @@ +"""Download pinned corpus files and record a frozen manifest. + +Upstream LFS pointers are unreliable: some declare a wrong size, some name an +oid that does not match the served blob. Files are hashed as served, every +pointer discrepancy is recorded in the manifest, and the corpus builder drops +documents whose annotations reference pages the served file lacks. +""" + +import argparse +import hashlib +import json +import subprocess +import sys +import urllib.error +import urllib.parse +import urllib.request +from datetime import datetime, timezone +from pathlib import Path + +from bench.corpus_sources import GIT_REPOS, SOURCES +from bench.sets_common import parse_json_line + +ROOT = Path(__file__).resolve().parents[1] +CORPUS = ROOT / "bench" / "corpus" +MANIFEST = CORPUS / "manifest.jsonl" +UA = {"User-Agent": "ely-eye-bench"} +OPENER = urllib.request.build_opener(urllib.request.HTTPSHandler()) + + +def http_get(url: str) -> tuple[bytes, dict[str, str]]: + if urllib.parse.urlsplit(url).scheme != "https" or urllib.parse.urlsplit(url).netloc != "huggingface.co": + raise SystemExit(f"refusing non-HF url: {url}") + try: + resp = OPENER.open(urllib.request.Request(url, headers=UA), timeout=300) # noqa: S310 + except urllib.error.HTTPError as exc: + raise SystemExit(f"HTTP {exc.code} {url}") from exc + except urllib.error.URLError as exc: + raise SystemExit(f"network {exc.reason} {url}") from exc + with resp: + return resp.read(), dict(resp.headers) + + +def sha256_bytes(data: bytes) -> str: + return hashlib.sha256(data).hexdigest() + + +def api(url: str) -> tuple[list, dict[str, str]]: + body, headers = http_get(url) + try: + payload = json.loads(body.decode("utf-8")) + except (json.JSONDecodeError, UnicodeDecodeError) as exc: + raise SystemExit(f"bad JSON from {url}: {exc}") from exc + return payload, headers + + +def next_link(headers: dict[str, str]) -> str | None: + header = headers.get("Link") + if not header: + return None + for part in header.split(","): + head, _, rel = part.partition(">") + if 'rel="next"' in rel: + return head.strip().strip("<>") + return None + + +def list_files(dataset: str, revision: str) -> list[dict]: + out: list[dict] = [] + url: str | None = f"https://huggingface.co/api/datasets/{dataset}/tree/{revision}?recursive=true" + while url: + batch, headers = api(url) + out += [f for f in batch if f["type"] == "file"] + url = next_link(headers) + return out + + +def file_digest(path: Path) -> tuple[int, str]: + digest = hashlib.sha256() + size = 0 + with path.open("rb") as handle: + for block in iter(lambda: handle.read(1 << 20), b""): + digest.update(block) + size += len(block) + return size, digest.hexdigest() + + +def download_to(url: str, dest: Path) -> tuple[int, str]: + dest.parent.mkdir(parents=True, exist_ok=True) + tmp = dest.with_suffix(dest.suffix + ".part") + digest = hashlib.sha256() + size = 0 + try: + resp = OPENER.open(urllib.request.Request(url, headers=UA), timeout=600) # noqa: S310 + except urllib.error.HTTPError as exc: + raise SystemExit(f"HTTP {exc.code} {url}") from exc + except urllib.error.URLError as exc: + raise SystemExit(f"network {exc.reason} {url}") from exc + try: + with resp, tmp.open("wb") as handle: + for block in iter(lambda: resp.read(1 << 20), b""): + digest.update(block) + size += len(block) + handle.write(block) + except OSError as exc: + tmp.unlink(missing_ok=True) + raise SystemExit(f"write failed for {url}: {exc}") from exc + tmp.replace(dest) + return size, digest.hexdigest() + + +def check_path(path: str) -> str: + if path.startswith("/") or ".." in Path(path).parts: + raise SystemExit(f"unsafe corpus path: {path}") + return path + + +def pointer_for(dataset: str, revision: str, path: str) -> tuple[str | None, int | None]: + raw, _ = http_get(f"https://huggingface.co/datasets/{dataset}/raw/{revision}/{path}") + if not raw.startswith(b"version https://git-lfs"): + return None, None + fields = dict( + line.decode().split(" ", 1) for line in raw.splitlines() if b" " in line + ) + oid = fields.get("oid", "").removeprefix("sha256:") or None + if "size" not in fields: + return oid, None + try: + size = int(fields["size"]) + except ValueError as exc: + raise SystemExit(f"bad LFS pointer size for {dataset}/{path}: {fields['size']!r}") from exc + return oid, size + + +def fetch_file(dataset: str, revision: str, path: str) -> dict: + check_path(path) + oid, declared = pointer_for(dataset, revision, path) + url = f"https://huggingface.co/datasets/{dataset}/resolve/{revision}/{path}" + target = CORPUS / dataset / path + if target.exists() and target.stat().st_size > 0: + size, actual = file_digest(target) + fresh = False + else: + size, actual = download_to(url, target) + fresh = True + return { + "dataset": dataset, + "revision": revision, + "path": path, + "url": url, + "bytes": size, + "sha256": actual, + "lfs_pointer_oid": oid, + "lfs_pointer_bytes": declared, + "oid_match": oid is None or actual == oid, + "newly_downloaded": fresh, + "retrieved_at": datetime.now(timezone.utc).isoformat(timespec="seconds"), + } + + +def fetch_git(repo: dict) -> dict: + name = repo["repo"].rsplit("/", 1)[-1] + target = CORPUS / "_git" / name + fresh = not target.exists() + if fresh: + subprocess.run( + ["git", "clone", "--depth", "1", "--branch", repo["ref"], repo["repo"], str(target)], + check=True, + stdout=subprocess.DEVNULL, + ) + head = subprocess.run( + ["git", "-C", str(target), "rev-parse", "HEAD"], check=True, capture_output=True, text=True + ).stdout.strip() + pinned = repo.get("sha") + if pinned and head != pinned: + raise SystemExit( + f"{repo['repo']}: checked-out HEAD {head} != pinned sha {pinned}; " + "re-fetch at the pinned commit or re-pin deliberately" + ) + tracked = [f for f in target.rglob("*") if f.is_file() and ".git" not in f.parts] + return { + "dataset": repo["repo"], + "revision": head, + "path": f"_git/{name}", + "url": f"{repo['repo']}/archive/{head}.tar.gz", + "bytes": sum(f.stat().st_size for f in tracked), + "sha256": head, + "lfs_pointer_oid": None, + "lfs_pointer_bytes": None, + "oid_match": True, + "newly_downloaded": fresh, + "retrieved_at": datetime.now(timezone.utc).isoformat(timespec="seconds"), + "file_count": len(tracked), + } + + +def existing_rows(exclude: set[str]) -> list[dict]: + """Keep previously fetched rows for datasets this run does not re-fetch.""" + if not MANIFEST.is_file(): + return [] + rows = [] + for number, line in enumerate(MANIFEST.read_text(encoding="utf-8").splitlines(), start=1): + if not line.strip(): + continue + row = parse_json_line(line, f"{MANIFEST}:{number}") + if row["dataset"] not in exclude: + rows.append(row) + return rows + + +def main() -> int: + ap = argparse.ArgumentParser() + ap.add_argument("--only", help="fetch one dataset id") + ap.add_argument("--skip-git", action="store_true") + args = ap.parse_args() + + CORPUS.mkdir(parents=True, exist_ok=True) + sources = [s for s in SOURCES if not args.only or s["dataset"] == args.only] + covered = {source["dataset"] for source in sources} + if not args.only and not args.skip_git: + covered |= {repo["repo"] for repo in GIT_REPOS} + rows = existing_rows(exclude=covered) + for source in sources: + if args.only and source["dataset"] != args.only: + continue + if source["files"] == "ALL": + names = [f["path"] for f in list_files(source["dataset"], source["revision"])] + else: + names = source["files"] + print(f"fetching {source['dataset']} ({len(names)} files)", flush=True) + for name in names: + row = fetch_file(source["dataset"], source["revision"], name) + flags = "" + if not row["oid_match"]: + flags += " <-- LFS oid mismatch" + if row["lfs_pointer_bytes"] not in (None, row["bytes"]): + flags += " <-- LFS size mismatch" + print(f" {row['bytes']:>10} {name}{flags}", flush=True) + rows.append( + { + **row, + "modality": source["modality"], + "license": source["license"], + "snapshot_at": source["snapshot_at"], + "annotation_source": source["annotation_source"], + } + ) + if not args.only and not args.skip_git: + for repo in GIT_REPOS: + print(f"cloning {repo['repo']}", flush=True) + rows.append( + { + **fetch_git(repo), + "modality": repo["modality"], + "license": "see-repo", + "snapshot_at": None, + "annotation_source": repo["annotation_source"], + } + ) + + rows.sort(key=lambda r: (r["dataset"], r["path"])) + MANIFEST.write_text("".join(json.dumps(r, sort_keys=True) + "\n" for r in rows), encoding="utf-8") + print(f"manifest: {len(rows)} files, {sum(r['bytes'] for r in rows) / 1e9:.2f} GB -> {MANIFEST}") + return 0 + + +if __name__ == "__main__": + sys.exit(main()) diff --git a/bench/sets/_cache/mmlongbench_pages.json b/bench/sets/_cache/mmlongbench_pages.json new file mode 100644 index 0000000..d638988 --- /dev/null +++ b/bench/sets/_cache/mmlongbench_pages.json @@ -0,0 +1,137 @@ +{ + "05-03-18-political-release.pdf": 17, + "0b85477387a9d0cc33fca0f4becaa0e5.pdf": 16, + "0e94b4197b10096b1f4c699701570fbf.pdf": 15, + "11-21-16-Updated-Post-Election-Release.pdf": 45, + "12-15-15-ISIS-and-terrorism-release-final.pdf": 28, + "2005.12872v3.pdf": 26, + "2021-Apple-Catalog.pdf": 55, + "2023.acl-long.386.pdf": 24, + "2023.findings-emnlp.248.pdf": 19, + "2024.ug.eprospectus.pdf": 27, + "2210.02442v1.pdf": 24, + "2303.05039v2.pdf": 9, + "2303.08559v2.pdf": 30, + "2305.13186v3.pdf": 27, + "2305.14160v4.pdf": 16, + "2306.05425v1.pdf": 22, + "2307.09288v2.pdf": 77, + "2309.17421v2.pdf": 166, + "2310.05634v2.pdf": 24, + "2310.07609v1.pdf": 10, + "2310.09158v1.pdf": 23, + "2311.16502v3.pdf": 117, + "2312.04350v3.pdf": 28, + "2312.09390v1.pdf": 49, + "2312.10997v5.pdf": 21, + "2401.18059v1.pdf": 23, + "2405.09818v1.pdf": 27, + "3276a5b991c49cf5f9a4af0f7d6fce67.pdf": 16, + "379f44022bb27aa53efd5d322c7b57bf.pdf": 17, + "3M_2018_10K.pdf": 160, + "52b3137455e7ca4df65021a200aef724.pdf": 112, + "698bba535087fa9a7f9009e172a7f763.pdf": 20, + "7c3f6204b3241f142f0f8eb8e1fefe7a.pdf": 15, + "8dfc21ec151fb9d3578fc32d5c4e5df9.pdf": 18, + "8e7c4cb542ad160f80fb3d795ada35d8.pdf": 17, + "91521110100M_4K_UHD_Display_User_Manual_V1.1.pdf": 40, + "936c0e2c2e6c8e0c07c51bfaf7fd0a83.pdf": 15, + "ACTIVISIONBLIZZARD_2019_10K.pdf": 198, + "ADOBE_2015_10K.pdf": 116, + "AMAZON_2017_10K.pdf": 85, + "BESTBUY_2023_10K.pdf": 75, + "BRO-GL-MMONEY.pdf": 16, + "Bergen-Brochure-en-2022-23.pdf": 46, + "COSTCO_2021_10K.pdf": 76, + "Campaign_038_Introducing_AC_Whitepaper_v5e.pdf": 28, + "DSA-278777.pdf": 21, + "GPL-Graduate-Studies-Professional-Learning-Brochure-Jul-2021.pdf": 17, + "Guide-for-international-students-web.pdf": 76, + "ISEP_student_handbook_2020.pdf": 24, + "Independents-Report.pdf": 23, + "Macbook_air.pdf": 76, + "NETFLIX_2015_10K.pdf": 72, + "NIKE_2021_10K.pdf": 109, + "NUS-Business-School-BBA-Brochure-2024.pdf": 24, + "NUS-FASS-Graduate-Guidebook-2021-small.pdf": 31, + "NYU_graduate.pdf": 30, + "PG_20.07.30_U.S.-Views-China_final.pdf": 22, + "PG_2020.03.09_US-Germany_FINAL.pdf": 21, + "PG_2020.05.21_International-Cooperation-COVID_FINAL.pdf": 26, + "PG_2021.03.04_US-Views-on-China_FINAL.pdf": 40, + "PH_2016.06.08_Economy-Final.pdf": 23, + "PIP_Seniors-and-Tech-Use_040314.pdf": 27, + "PI_2017.10.04_Automation_FINAL.pdf": 61, + "PI_2018.11.19_algorithms_FINAL.pdf": 30, + "PP_2019.01.17_Trump-economy_FINAL2.pdf": 28, + "PP_2020.08.06_COVID-19-Restrictions_FINAL-1.pdf": 26, + "PP_2021.04.22_voting-access_REPORT.pdf": 24, + "PRE_2022.09.29_NSL-politics_REPORT.pdf": 108, + "PS_2018.01.09_STEM_FINAL.pdf": 105, + "PWC_opportunity_of_lifetime.pdf": 18, + "Pew-Research-Center_Hispanic-Identity-Report_12.20.2017.pdf": 32, + "RAR.pdf": 28, + "SAO-StudentSupport_Guidebook-Content.pdf": 44, + "STEPBACK.pdf": 36, + "Sinopolis-Chengdu.pdf": 96, + "SnapNTell.pdf": 20, + "StudentSupport_Guidebook.pdf": 44, + "User_Manual_1500S_Classic_EN.pdf": 108, + "a4f3ced0696009fec3179f493e4f28c4.pdf": 17, + "a5879805d70c854ea4361e43a84e3bb2.pdf": 15, + "afe620b9beac86c1027b96d31d396407.pdf": 20, + "amb-siteaudits-ds15-150204174043-conversion-gate01_95.pdf": 41, + "asdaaburson-marstellerarabyouthsurvey2014-140407100615-phpapp01_95.pdf": 45, + "avalaunchpresentationsthatkickasteriskv3copy-150318114804-conversion-gate01_95.pdf": 27, + "b3m5kaeqm2w8n4bwcesw-140602121350-phpapp02_95.pdf": 21, + "bariumswallowpresentation-090810084400-phpapp01_95.pdf": 77, + "bdf54dxa.pdf": 54, + "bigdatatrends-120723191058-phpapp02_95.pdf": 53, + "c31e6580d0175ab3f9d99d1ff0bfa000.pdf": 20, + "caltraincapacitymountainview1-150701205750-lva1-app6891_95.pdf": 31, + "camry_ebrochure.pdf": 26, + "catvsdogdlpycon15se-150512122612-lva1-app6891_95.pdf": 68, + "chapter8-geneticscompatibilitymode-141214140247-conversion-gate02_95.pdf": 78, + "competitiveoutcomes-091006065143-phpapp01_95.pdf": 48, + "csewt7zsecmmbzjufbyx-signature-24d91a254426c21c3079384270e1f138dc43a271cfe15d6d520d68205855b2a3-poli-150306115347-conversion-gate01_95.pdf": 65, + "ddoseattle-150627210357-lva1-app6891_95.pdf": 63, + "digitalmeasurementframework22feb2011v6novideo-110221233835-phpapp01_95.pdf": 196, + "disciplined-agile-business-analysis-160218012713_95.pdf": 41, + "dr-vorapptchapter1emissionsources-121120210508-phpapp02_95.pdf": 196, + "e639029d16094ea71d964e2fb953952b.pdf": 20, + "e79deb02a0c0e87511080836c5d4347b.pdf": 17, + "earlybird-110722143746-phpapp02_95.pdf": 41, + "earthlinkweb-150213112111-conversion-gate02_95.pdf": 44, + "ecommerceopportunityindia-141124010546-conversion-gate01_95.pdf": 28, + "edb88a99670417f64a6b719646aed326.pdf": 20, + "efd88e41c5f2606c57929cac6c1c0605.pdf": 20, + "efis-140411041451-phpapp01_95.pdf": 32, + "f1f5242528411b262be447e61e2eb10f.pdf": 20, + "f86d073b0d735ac873a65d906ba82758.pdf": 20, + "f8d3a162ab9507e021d83dd109118b60.pdf": 17, + "fd76bbefe469561966e5387aa709c482.pdf": 17, + "fdac8d1e9ef56519371df7e6532df27d.pdf": 20, + "finalmediafindingspdf-141228031149-conversion-gate02_95.pdf": 44, + "finalpresentationdeck-whatwhyhowofcertificationsocial-160324220748_95.pdf": 35, + "formwork-150318073913-conversion-gate01_95.pdf": 60, + "germanwingsdigitalcrisisanalysis-150403064828-conversion-gate01_95.pdf": 23, + "guojixueshengshenghuozhinanyingwen9.1.pdf": 40, + "honor_watch_gs_pro.pdf": 42, + "indonesiamobilemarketresearch-ag-150106055934-conversion-gate02_95.pdf": 38, + "measuringsuccessonfacebooktwitterlinkedin-160317142140_95.pdf": 25, + "mi_phone.pdf": 30, + "mmdetection-readthedocs-io-en-v2.18.0.pdf": 468, + "nielsen2015musicbizpresentation-final-150526143534-lva1-app6891_95.pdf": 36, + "nova_y70.pdf": 45, + "obs-productdesc-en.pdf": 65, + "owners-manual-2170416.pdf": 32, + "q1-2023-bilibili-inc-investor-presentation.pdf": 27, + "reportq32015-151009093138-lva1-app6891_95.pdf": 34, + "san-francisco-11-contents.pdf": 40, + "stereo_headset.pdf": 151, + "t480_ug_en.pdf": 170, + "tacl_a_00660.pdf": 23, + "transform-software-delivery-with-valueedge-brochure.pdf": 20, + "watch_d.pdf": 27, + "welcome-to-nus.pdf": 24 +} diff --git a/bench/sets/code.jsonl b/bench/sets/code.jsonl new file mode 100644 index 0000000..cf56590 --- /dev/null +++ b/bench/sets/code.jsonl @@ -0,0 +1,2474 @@ +{"answer": "csn:20tab/twentytab-treeeditor:treeeditor/admin.py:TreeEditor._collect_editable_booleans#20240", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "treeeditor/admin.py", "symbol": "TreeEditor._collect_editable_booleans"}, "source_id": "csn:20tab/twentytab-treeeditor:treeeditor/admin.py:TreeEditor._collect_editable_booleans#20240"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/20tab/twentytab-treeeditor/blob/f89d459b1348961880cd488df95690e68529f96b/treeeditor/admin.py#L215-L240", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:20tab-twentytab-treeeditor:TreeEditor._collect_editable_booleans-b65951cde5", "question": "Collect all fields marked as editable booleans. We do not\n want the user to be able to edit arbitrary fields by crafting\n an AJAX request by hand.", "scope": {"holder": "20tab-twentytab-treeeditor", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:20tab/twentytab-treeeditor:treeeditor/admin.py:TreeEditor.has_delete_permission#20244", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "treeeditor/admin.py", "symbol": "TreeEditor.has_delete_permission"}, "source_id": "csn:20tab/twentytab-treeeditor:treeeditor/admin.py:TreeEditor.has_delete_permission#20244"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/20tab/twentytab-treeeditor/blob/f89d459b1348961880cd488df95690e68529f96b/treeeditor/admin.py#L363-L374", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:20tab-twentytab-treeeditor:TreeEditor.has_delete_permission-ca0cf22fa4", "question": "Implement a lookup for object level permissions. Basically the same as\n ModelAdmin.has_delete_permission, but also passes the obj parameter in.", "scope": {"holder": "20tab-twentytab-treeeditor", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:3DLIRIOUS/MeshLabXML:meshlabxml/mlx.py:FilterScript.add_layer#6360", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "meshlabxml/mlx.py", "symbol": "FilterScript.add_layer"}, "source_id": "csn:3DLIRIOUS/MeshLabXML:meshlabxml/mlx.py:FilterScript.add_layer#6360"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/3DLIRIOUS/MeshLabXML/blob/177cce21e92baca500f56a932d66bd9a33257af8/meshlabxml/mlx.py#L181-L191", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:3DLIRIOUS-MeshLabXML:FilterScript.add_layer-6d01968d4e", "question": "Add new mesh layer to the end of the stack\n\n Args:\n label (str): new label for the mesh layer\n change_layer (bool): change to the newly created layer", "scope": {"holder": "3DLIRIOUS-MeshLabXML", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:3DLIRIOUS/MeshLabXML:meshlabxml/compute.py:measure_geometry#6387", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "meshlabxml/compute.py", "symbol": "measure_geometry"}, "source_id": "csn:3DLIRIOUS/MeshLabXML:meshlabxml/compute.py:measure_geometry#6387"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/3DLIRIOUS/MeshLabXML/blob/177cce21e92baca500f56a932d66bd9a33257af8/meshlabxml/compute.py#L117-L143", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:3DLIRIOUS-MeshLabXML:measure_geometry-ec89a54ba7", "question": "Compute a set of geometric measures of a mesh/pointcloud.\n\n Bounding box extents and diagonal, principal axis, thin shell barycenter\n (mesh only), vertex barycenter and quality-weighted barycenter (pointcloud\n only), surface area (mesh only), volume (closed mesh) and Inertia tensor\n Matrix (closed mesh).\n\n Args:\n script: the mlx.FilterScript object or script filename to write\n the filter to.\n\n Layer stack:\n No impacts\n\n MeshLab versions:\n 2016.12\n 1.3.4BETA\n\n Bugs:\n Bounding box extents not computed correctly for some volumes", "scope": {"holder": "3DLIRIOUS-MeshLabXML", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:4degrees/riffle:setup.py:BuildResources.finalize_options#15487", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "setup.py", "symbol": "BuildResources.finalize_options"}, "source_id": "csn:4degrees/riffle:setup.py:BuildResources.finalize_options#15487"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/4degrees/riffle/blob/e5a0d908df8c93ff1ee7abdda8875fd1667df53d/setup.py#L48-L51", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:4degrees-riffle:BuildResources.finalize_options-d0399ca77d", "question": "Finalize options to be used.", "scope": {"holder": "4degrees-riffle", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:4degrees/riffle:source/riffle/browser.py:FilesystemBrowser._onNavigate#15483", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "source/riffle/browser.py", "symbol": "FilesystemBrowser._onNavigate"}, "source_id": "csn:4degrees/riffle:source/riffle/browser.py:FilesystemBrowser._onNavigate#15483"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/4degrees/riffle/blob/e5a0d908df8c93ff1ee7abdda8875fd1667df53d/source/riffle/browser.py#L146-L151", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:4degrees-riffle:FilesystemBrowser._onNavigate-3b6ef681ea", "question": "Handle selection of path segment.", "scope": {"holder": "4degrees-riffle", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:5j9/wikitextparser:wikitextparser/_wikitext.py:SubWikiText.ancestors#7911", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "wikitextparser/_wikitext.py", "symbol": "SubWikiText.ancestors"}, "source_id": "csn:5j9/wikitextparser:wikitextparser/_wikitext.py:SubWikiText.ancestors#7911"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/5j9/wikitextparser/blob/1347425814361d7955342c53212edbb27f0ff4b5/wikitextparser/_wikitext.py#L1065-L1088", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:5j9-wikitextparser:SubWikiText.ancestors-a2cb35e759", "question": "Return the ancestors of the current node.\n\n :param type_: the type of the desired ancestors as a string.\n Currently the following types are supported: {Template,\n ParserFunction, WikiLink, Comment, Parameter, ExtensionTag}.\n The default is None and means all the ancestors of any type above.", "scope": {"holder": "5j9-wikitextparser", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:5j9/wikitextparser:wikitextparser/_wikitext.py:WikiText._pp_type_to_spans#7897", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "wikitextparser/_wikitext.py", "symbol": "WikiText._pp_type_to_spans"}, "source_id": "csn:5j9/wikitextparser:wikitextparser/_wikitext.py:WikiText._pp_type_to_spans#7897"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/5j9/wikitextparser/blob/1347425814361d7955342c53212edbb27f0ff4b5/wikitextparser/_wikitext.py#L481-L494", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:5j9-wikitextparser:WikiText._pp_type_to_spans-b5c66d196f", "question": "Create the arguments for the parse function used in pformat method.\n\n Only return sub-spans and change the them to fit the new scope, i.e\n self.string.", "scope": {"holder": "5j9-wikitextparser", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:5monkeys/django-bananas:bananas/admin/api/mixins.py:BananasAPI.get_view_name#12384", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "bananas/admin/api/mixins.py", "symbol": "BananasAPI.get_view_name"}, "source_id": "csn:5monkeys/django-bananas:bananas/admin/api/mixins.py:BananasAPI.get_view_name#12384"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/5monkeys/django-bananas/blob/cfd318c737f6c4580036c13d2acf32bca96654bf/bananas/admin/api/mixins.py#L90-L116", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:5monkeys-django-bananas:BananasAPI.get_view_name-a09ab63a5e", "question": "Get or generate human readable view name.\n Extended version from DRF to support usage from both class and instance.", "scope": {"holder": "5monkeys-django-bananas", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:5monkeys/django-bananas:bananas/url.py:get_engine#12390", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "bananas/url.py", "symbol": "get_engine"}, "source_id": "csn:5monkeys/django-bananas:bananas/url.py:get_engine#12390"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/5monkeys/django-bananas/blob/cfd318c737f6c4580036c13d2acf32bca96654bf/bananas/url.py#L92-L134", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:5monkeys-django-bananas:get_engine-19ff20a9db", "question": "Perform a lookup in _ENGINE_MAPPING using engine_string.\n\n :param scheme: '+'-separated string Maximum of 2 parts,\n i.e \"postgres+psycopg\" is OK, \"postgres+psycopg2+postgis\" is NOT OK.\n :return: Engine string", "scope": {"holder": "5monkeys-django-bananas", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:AartGoossens/goldencheetahlib:goldencheetahlib/client.py:GoldenCheetahClient._get_request#19503", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "goldencheetahlib/client.py", "symbol": "GoldenCheetahClient._get_request"}, "source_id": "csn:AartGoossens/goldencheetahlib:goldencheetahlib/client.py:GoldenCheetahClient._get_request#19503"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/AartGoossens/goldencheetahlib/blob/ebe57de7d94280674c8440a81f53ac02f0b4eb43/goldencheetahlib/client.py#L155-L181", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:AartGoossens-goldencheetahlib:GoldenCheetahClient._get_request-4adc3e6124", "question": "Do actual GET request to GC REST API\n Also validates responses.\n\n Keyword arguments:\n endpoint -- full endpoint for GET request", "scope": {"holder": "AartGoossens-goldencheetahlib", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:AartGoossens/goldencheetahlib:goldencheetahlib/client.py:GoldenCheetahClient._request_activity_list#19499", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "goldencheetahlib/client.py", "symbol": "GoldenCheetahClient._request_activity_list"}, "source_id": "csn:AartGoossens/goldencheetahlib:goldencheetahlib/client.py:GoldenCheetahClient._request_activity_list#19499"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/AartGoossens/goldencheetahlib/blob/ebe57de7d94280674c8440a81f53ac02f0b4eb43/goldencheetahlib/client.py#L85-L110", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:AartGoossens-goldencheetahlib:GoldenCheetahClient._request_activity_list-a30b2174ef", "question": "Actually do the request for activity list\n This call is slow and therefore this method is memory cached.\n\n Keyword arguments:\n athlete -- Full name of athlete", "scope": {"holder": "AartGoossens-goldencheetahlib", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:AkihikoITOH/capybara:capybara/virtualenv/lib/python2.7/site-packages/pip/_vendor/_markerlib/markers.py:ASTWhitelist.visit#15862", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "capybara/virtualenv/lib/python2.7/site-packages/pip/_vendor/_markerlib/markers.py", "symbol": "ASTWhitelist.visit"}, "source_id": "csn:AkihikoITOH/capybara:capybara/virtualenv/lib/python2.7/site-packages/pip/_vendor/_markerlib/markers.py:ASTWhitelist.visit#15862"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/AkihikoITOH/capybara/blob/e86c2173ea386654f4ae061148e8fbe3f25e715c/capybara/virtualenv/lib/python2.7/site-packages/pip/_vendor/_markerlib/markers.py#L70-L76", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:AkihikoITOH-capybara:ASTWhitelist.visit-d73207e9cb", "question": "Ensure statement only contains allowed nodes.", "scope": {"holder": "AkihikoITOH-capybara", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:AkihikoITOH/capybara:capybara/virtualenv/lib/python2.7/site-packages/pip/vcs/__init__.py:VersionControl.run_command#16187", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "capybara/virtualenv/lib/python2.7/site-packages/pip/vcs/__init__.py", "symbol": "VersionControl.run_command"}, "source_id": "csn:AkihikoITOH/capybara:capybara/virtualenv/lib/python2.7/site-packages/pip/vcs/__init__.py:VersionControl.run_command#16187"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/AkihikoITOH/capybara/blob/e86c2173ea386654f4ae061148e8fbe3f25e715c/capybara/virtualenv/lib/python2.7/site-packages/pip/vcs/__init__.py#L301-L321", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:AkihikoITOH-capybara:VersionControl.run_command-d456583f2b", "question": "Run a VCS subcommand\n This is simply a wrapper around call_subprocess that adds the VCS\n command name, and checks that the VCS is available", "scope": {"holder": "AkihikoITOH-capybara", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Alignak-monitoring-contrib/alignak-backend-client:alignak_backend_client/client.py:Backend.decode#14960", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "alignak_backend_client/client.py", "symbol": "Backend.decode"}, "source_id": "csn:Alignak-monitoring-contrib/alignak-backend-client:alignak_backend_client/client.py:Backend.decode#14960"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Alignak-monitoring-contrib/alignak-backend-client/blob/1e21f6ce703e66984d1f9b20fe7866460ab50b39/alignak_backend_client/client.py#L199-L221", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Alignak-monitoring-contrib-alignak-backend-client:Backend.decode-f283cc66b1", "question": "Decodes and returns the response as JSON (dict) or raise BackendException\n :param response: requests.response object\n :return: dict", "scope": {"holder": "Alignak-monitoring-contrib-alignak-backend-client", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Alignak-monitoring-contrib/alignak-backend-client:alignak_backend_client/client.py:Backend.set_token#14961", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "alignak_backend_client/client.py", "symbol": "Backend.set_token"}, "source_id": "csn:Alignak-monitoring-contrib/alignak-backend-client:alignak_backend_client/client.py:Backend.set_token#14961"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Alignak-monitoring-contrib/alignak-backend-client/blob/1e21f6ce703e66984d1f9b20fe7866460ab50b39/alignak_backend_client/client.py#L223-L238", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Alignak-monitoring-contrib-alignak-backend-client:Backend.set_token-e117ab4734", "question": "Set token in authentification for next requests\n :param token: str. token to set in auth. If None, reinit auth", "scope": {"holder": "Alignak-monitoring-contrib-alignak-backend-client", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:AnthonyMRios/pymetamap:pymetamap/SubprocessBackend.py:SubprocessBackend.extract_concepts#9010", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pymetamap/SubprocessBackend.py", "symbol": "SubprocessBackend.extract_concepts"}, "source_id": "csn:AnthonyMRios/pymetamap:pymetamap/SubprocessBackend.py:SubprocessBackend.extract_concepts#9010"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/AnthonyMRios/pymetamap/blob/01f35802ad13f2eb7dff033e66e5445e8f5bc039/pymetamap/SubprocessBackend.py#L27-L175", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:AnthonyMRios-pymetamap:SubprocessBackend.extract_concepts-2ae2558c37", "question": "extract_concepts takes a list of sentences and ids(optional)\n then returns a list of Concept objects extracted via\n MetaMap.\n\n Supported Options:\n Composite Phrase -Q\n Word Sense Disambiguation -y\n use strict model -A\n use relaxed model -C\n allow large N -l\n allow overmatches -o\n allow concept gaps -g\n term processing -z\n No Derivational Variants -d\n All Derivational Variants -D\n Ignore Word Order -i\n Allow Acronym Variants -a\n Unique Acronym Variants -u\n Prefer Multiple Concepts -Y\n Ignore Stop Phrases -K\n Compute All Mappings -b\n MM Data Version -V\n Exclude Sources -e\n Restrict to Sources -R\n Restrict to Semantic Types -J\n Exclude Semantic Types -k\n \n\n For information about the available options visit\n http://metamap.nlm.nih.gov/.\n\n Note: If an error is encountered the process will be closed\n and whatever was processed, if anything, will be\n returned along with the error found.", "scope": {"holder": "AnthonyMRios-pymetamap", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:AnthonyMRios/pymetamap:pymetamap/SubprocessBackendLite.py:SubprocessBackendLite.extract_concepts#9011", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pymetamap/SubprocessBackendLite.py", "symbol": "SubprocessBackendLite.extract_concepts"}, "source_id": "csn:AnthonyMRios/pymetamap:pymetamap/SubprocessBackendLite.py:SubprocessBackendLite.extract_concepts#9011"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/AnthonyMRios/pymetamap/blob/01f35802ad13f2eb7dff033e66e5445e8f5bc039/pymetamap/SubprocessBackendLite.py#L27-L115", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:AnthonyMRios-pymetamap:SubprocessBackendLite.extract_concepts-d754c27ab1", "question": "extract_concepts takes a list of sentences and ids(optional)\n then returns a list of Concept objects extracted via\n MetaMapLite.\n\n Supported Options:\n Restrict to Semantic Types --restrict_to_sts\n Restrict to Sources --restrict_to_sources\n\n For information about the available options visit\n http://metamap.nlm.nih.gov/.\n\n Note: If an error is encountered the process will be closed\n and whatever was processed, if anything, will be\n returned along with the error found.", "scope": {"holder": "AnthonyMRios-pymetamap", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Azure/Azure-MachineLearning-ClientLibrary-Python:azureml/__init__.py:Datasets.add_from_dataframe#7275", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "azureml/__init__.py", "symbol": "Datasets.add_from_dataframe"}, "source_id": "csn:Azure/Azure-MachineLearning-ClientLibrary-Python:azureml/__init__.py:Datasets.add_from_dataframe#7275"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Azure/Azure-MachineLearning-ClientLibrary-Python/blob/d1211b289747671898eb063013e0dc53d3c80acd/azureml/__init__.py#L465-L507", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Azure-Azure-MachineLearning-ClientLibrary-Python:Datasets.add_from_dataframe-838a58a2f2", "question": "Serialize the specified DataFrame and upload it as a new dataset.\n\n Parameters\n ----------\n dataframe : pandas.DataFrame\n Data to serialize.\n data_type_id : str\n Format to serialize to.\n Supported formats are:\n 'PlainText'\n 'GenericCSV'\n 'GenericTSV'\n 'GenericCSVNoHeader'\n 'GenericTSVNoHeader'\n See the azureml.DataTypeIds class for constants.\n name : str\n Name for the new dataset.\n description : str\n Description for the new dataset.\n\n Returns\n -------\n SourceDataset\n Dataset that was just created.\n Use open(), read_as_binary(), read_as_text() or to_dataframe() on\n the dataset object to get its contents as a stream, bytes, str or\n pandas DataFrame.", "scope": {"holder": "Azure-Azure-MachineLearning-ClientLibrary-Python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Azure/Azure-MachineLearning-ClientLibrary-Python:azureml/http.py:_RestClient.get_datasets#7283", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "azureml/http.py", "symbol": "_RestClient.get_datasets"}, "source_id": "csn:Azure/Azure-MachineLearning-ClientLibrary-Python:azureml/http.py:_RestClient.get_datasets#7283"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Azure/Azure-MachineLearning-ClientLibrary-Python/blob/d1211b289747671898eb063013e0dc53d3c80acd/azureml/http.py#L70-L73", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Azure-Azure-MachineLearning-ClientLibrary-Python:_RestClient.get_datasets-a426b60d85", "question": "Runs HTTP GET request to retrieve the list of datasets.", "scope": {"holder": "Azure-Azure-MachineLearning-ClientLibrary-Python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Azure/azure-sdk-for-python:azure-mgmt-alertsmanagement/azure/mgmt/alertsmanagement/operations/alerts_operations.py:AlertsOperations.get_all#2443", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "azure-mgmt-alertsmanagement/azure/mgmt/alertsmanagement/operations/alerts_operations.py", "symbol": "AlertsOperations.get_all"}, "source_id": "csn:Azure/azure-sdk-for-python:azure-mgmt-alertsmanagement/azure/mgmt/alertsmanagement/operations/alerts_operations.py:AlertsOperations.get_all#2443"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Azure/azure-sdk-for-python/blob/d7306fde32f60a293a7567678692bdad31e4b667/azure-mgmt-alertsmanagement/azure/mgmt/alertsmanagement/operations/alerts_operations.py#L39-L210", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Azure-azure-sdk-for-python:AlertsOperations.get_all-258f141cc1", "question": "List all the existing alerts, where the results can be selective by\n passing multiple filter parameters including time range and sorted on\n specific fields. .\n\n :param target_resource: Filter by target resource( which is full ARM\n ID) Default value is select all.\n :type target_resource: str\n :param target_resource_type: Filter by target resource type. Default\n value is select all.\n :type target_resource_type: str\n :param target_resource_group: Filter by target resource group name.\n Default value is select all.\n :type target_resource_group: str\n :param monitor_service: Filter by monitor service which is the source\n of the alert instance. Default value is select all. Possible values\n include: 'Application Insights', 'ActivityLog Administrative',\n 'ActivityLog Security', 'ActivityLog Recommendation', 'ActivityLog\n Policy', 'ActivityLog Autoscale', 'Log Analytics', 'Nagios',\n 'Platform', 'SCOM', 'ServiceHealth', 'SmartDetector', 'VM Insights',\n 'Zabbix'\n :type monitor_service: str or\n ~azure.mgmt.alertsmanagement.models.MonitorService\n :param monitor_condition: Filter by monitor condition which is the\n state of the monitor(alertRule) at monitor service. Default value is\n to select all. Possible values include: 'Fired', 'Resolved'\n :type monitor_condition: str or\n ~azure.mgmt.alertsmanagement.models.MonitorCondition\n :param severity: Filter by severity. Defaut value is select all.\n Possible values include: 'Sev0', 'Sev1', 'Sev2', 'Sev3', 'Sev4'\n :type severity: str or ~azure.mgmt.alertsmanagement.models.Severity\n :param alert_state: Filter by state of the alert instance. Default\n value is to select all. Possible values include: 'New',\n 'Acknowledged', 'Closed'\n :type alert_state: str or\n ~azure.mgmt.alertsmanagement.models.AlertState\n :param alert_rule: Filter by alert rule(monitor) which fired alert\n instance. Default value is to select all.\n :type alert_rule: str\n :param smart_group_id: Filter the alerts list by the Smart Group Id.\n Default value is none.\n :type smart_group_id: str\n :param include_context: Include context which has data contextual to\n the monitor service. Default value is false'\n :type include_context: bool\n :param include_egress_config: Include egress config which would be\n used for displaying the content in portal. Default value is 'false'.\n :type include_egress_config: bool\n :param page_count: Determines number of alerts returned per page in\n response. Permissible value is between 1 to 250. When the\n \"includeContent\" filter is selected, maximum value allowed is 25.\n Default value is 25.\n :type page_count: int\n :param sort_by: Sort the query results by input field, Default value\n is 'lastModifiedDateTime'. Possible values include: 'name',\n 'severity', 'alertState', 'monitorCondition', 'targetResource',\n 'targetResourceName', 'targetResourceGroup', 'targetResourceType',\n 'startDateTime', 'lastModifiedDateTime'\n :type sort_by: str or\n ~azure.mgmt.alertsmanagement.models.AlertsSortByFields\n :param sort_order: Sort the query results order in either ascending or\n descending. Default value is 'desc' for time fields and 'asc' for\n others. Possible values include: 'asc', 'desc'\n :type sort_order: str\n :param select: This filter allows to selection of the fields(comma\n seperated) which would be part of the the essential section. This\n would allow to project only the required fields rather than getting\n entire content. Default is to fetch all the fields in the essentials\n section.\n :type select: str\n :param time_range: Filter by time range by below listed values.\n Default value is 1 day. Possible values include: '1h', '1d', '7d',\n '30d'\n :type time_range: str or ~azure.mgmt.alertsmanagement.models.TimeRange\n :param custom_time_range: Filter by custom time range in the format\n / where time is in (ISO-8601 format)'.\n Permissible values is within 30 days from query time. Either\n timeRange or customTimeRange could be used but not both. Default is\n none.\n :type custom_time_range: str\n :param dict custom_headers: headers that will be added to the request\n :param bool raw: returns the direct response alongside the\n deserialized response\n :param operation_config: :ref:`Operation configuration\n overrides`.\n :return: An iterator like instance of Alert\n :rtype:\n ~azure.mgmt.alertsmanagement.models.AlertPaged[~azure.mgmt.alertsmanagement.models.Alert]\n :raises:\n :class:`ErrorResponseException`", "scope": {"holder": "Azure-azure-sdk-for-python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Azure/azure-sdk-for-python:azure-servicebus/azure/servicebus/control_client/servicebusservice.py:ServiceBusService.with_filter#2710", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "azure-servicebus/azure/servicebus/control_client/servicebusservice.py", "symbol": "ServiceBusService.with_filter"}, "source_id": "csn:Azure/azure-sdk-for-python:azure-servicebus/azure/servicebus/control_client/servicebusservice.py:ServiceBusService.with_filter#2710"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Azure/azure-sdk-for-python/blob/d7306fde32f60a293a7567678692bdad31e4b667/azure-servicebus/azure/servicebus/control_client/servicebusservice.py#L182-L201", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Azure-azure-sdk-for-python:ServiceBusService.with_filter-8200cbb233", "question": "Returns a new service which will process requests with the specified\n filter. Filtering operations can include logging, automatic retrying,\n etc... The filter is a lambda which receives the HTTPRequest and\n another lambda. The filter can perform any pre-processing on the\n request, pass it off to the next lambda, and then perform any\n post-processing on the response.", "scope": {"holder": "Azure-azure-sdk-for-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:BD2KGenomics/toil-lib:src/toil_lib/abstractPipelineWrapper.py:AbstractPipelineWrapper.__get_empty_config#15688", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/toil_lib/abstractPipelineWrapper.py", "symbol": "AbstractPipelineWrapper.__get_empty_config"}, "source_id": "csn:BD2KGenomics/toil-lib:src/toil_lib/abstractPipelineWrapper.py:AbstractPipelineWrapper.__get_empty_config#15688"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/BD2KGenomics/toil-lib/blob/022a615fc3dc98fc1aaa7bfd232409962ca44fbd/src/toil_lib/abstractPipelineWrapper.py#L123-L132", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:BD2KGenomics-toil-lib:AbstractPipelineWrapper.__get_empty_config-68bffcf2f5", "question": "Returns the config file contents as a string. The config file is generated and then deleted.", "scope": {"holder": "BD2KGenomics-toil-lib", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:BD2KGenomics/toil-lib:src/toil_lib/tools/variant_manipulation.py:gatk_select_variants#15736", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/toil_lib/tools/variant_manipulation.py", "symbol": "gatk_select_variants"}, "source_id": "csn:BD2KGenomics/toil-lib:src/toil_lib/tools/variant_manipulation.py:gatk_select_variants#15736"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/BD2KGenomics/toil-lib/blob/022a615fc3dc98fc1aaa7bfd232409962ca44fbd/src/toil_lib/tools/variant_manipulation.py#L7-L44", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:BD2KGenomics-toil-lib:gatk_select_variants-e9c7ecf3de", "question": "Isolates a particular variant type from a VCF file using GATK SelectVariants\n\n :param JobFunctionWrappingJob job: passed automatically by Toil\n :param str mode: variant type (i.e. SNP or INDEL)\n :param str vcf_id: FileStoreID for input VCF file\n :param str ref_fasta: FileStoreID for reference genome fasta\n :param str ref_fai: FileStoreID for reference genome index file\n :param str ref_dict: FileStoreID for reference genome sequence dictionary file\n :return: FileStoreID for filtered VCF\n :rtype: str", "scope": {"holder": "BD2KGenomics-toil-lib", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:BlueDragonX/detach:detach.py:Detach._close_fd#15517", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "detach.py", "symbol": "Detach._close_fd"}, "source_id": "csn:BlueDragonX/detach:detach.py:Detach._close_fd#15517"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/BlueDragonX/detach/blob/e2e5a1076e19f508baf3ffb2b586a75934fbae28/detach.py#L50-L57", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:BlueDragonX-detach:Detach._close_fd-f10b509847", "question": "Close a file descriptor if it is open.", "scope": {"holder": "BlueDragonX-detach", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:BlueDragonX/detach:detach.py:Detach._close_open_fds#15518", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "detach.py", "symbol": "Detach._close_open_fds"}, "source_id": "csn:BlueDragonX/detach:detach.py:Detach._close_open_fds#15518"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/BlueDragonX/detach/blob/e2e5a1076e19f508baf3ffb2b586a75934fbae28/detach.py#L59-L64", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:BlueDragonX-detach:Detach._close_open_fds-d32fd2eb91", "question": "Close open file descriptors.", "scope": {"holder": "BlueDragonX-detach", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:CITGuru/PyInquirer:PyInquirer/color_print.py:_print_token_factory#6173", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "PyInquirer/color_print.py", "symbol": "_print_token_factory"}, "source_id": "csn:CITGuru/PyInquirer:PyInquirer/color_print.py:_print_token_factory#6173"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/CITGuru/PyInquirer/blob/10d53723b36ebc7bba311457ec4afd9747a5c777/PyInquirer/color_print.py#L10-L27", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:CITGuru-PyInquirer:_print_token_factory-163ec15cd2", "question": "Internal helper to provide color names.", "scope": {"holder": "CITGuru-PyInquirer", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:CLARIAH/iribaker:iribaker/__init__.py:to_iri#15619", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "iribaker/__init__.py", "symbol": "to_iri"}, "source_id": "csn:CLARIAH/iribaker:iribaker/__init__.py:to_iri#15619"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/CLARIAH/iribaker/blob/47d2e8e95472353769962fde7626881f53429379/iribaker/__init__.py#L17-L81", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:CLARIAH-iribaker:to_iri-573c50f2b2", "question": "Safely quotes an IRI in a way that is resilient to unicode and incorrect\n arguments (checks for RFC 3987 compliance and falls back to percent encoding)", "scope": {"holder": "CLARIAH-iribaker", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Chilipp/sphinx-nbexamples:sphinx_nbexamples/__init__.py:Gallery.process_directories#14846", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sphinx_nbexamples/__init__.py", "symbol": "Gallery.process_directories"}, "source_id": "csn:Chilipp/sphinx-nbexamples:sphinx_nbexamples/__init__.py:Gallery.process_directories#14846"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Chilipp/sphinx-nbexamples/blob/08e0319ff3c70f8a931dfa8890caf48add4d0470/sphinx_nbexamples/__init__.py#L778-L784", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Chilipp-sphinx-nbexamples:Gallery.process_directories-95c4e3c0cb", "question": "Create the rst files from the input directories in the\n :attr:`in_dir` attribute", "scope": {"holder": "Chilipp-sphinx-nbexamples", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Chilipp/sphinx-nbexamples:sphinx_nbexamples/__init__.py:NotebookProcessor.data_download#14840", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sphinx_nbexamples/__init__.py", "symbol": "NotebookProcessor.data_download"}, "source_id": "csn:Chilipp/sphinx-nbexamples:sphinx_nbexamples/__init__.py:NotebookProcessor.data_download#14840"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Chilipp/sphinx-nbexamples/blob/08e0319ff3c70f8a931dfa8890caf48add4d0470/sphinx_nbexamples/__init__.py#L486-L492", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Chilipp-sphinx-nbexamples:NotebookProcessor.data_download-8fa730d769", "question": "Create the rst string to download supplementary data", "scope": {"holder": "Chilipp-sphinx-nbexamples", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:ChrisBeaumont/soupy:soupy.py:Collection.dictzip#11946", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "soupy.py", "symbol": "Collection.dictzip"}, "source_id": "csn:ChrisBeaumont/soupy:soupy.py:Collection.dictzip#11946"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ChrisBeaumont/soupy/blob/795f2f61f711f574d5218fc8a3375d02bda1104f/soupy.py#L708-L723", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ChrisBeaumont-soupy:Collection.dictzip-4c69966839", "question": "Turn this collection into a Scalar(dict), by zipping keys and items.\n\n Parameters:\n\n keys: list or Collection of NavigableStrings\n The keys of the dictionary\n\n Examples:\n\n >>> c = Collection([Scalar(1), Scalar(2)])\n >>> c.dictzip(['a', 'b']).val() == {'a': 1, 'b': 2}\n True", "scope": {"holder": "ChrisBeaumont-soupy", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:ChrisBeaumont/soupy:soupy.py:Node.find_next_sibling#11948", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "soupy.py", "symbol": "Node.find_next_sibling"}, "source_id": "csn:ChrisBeaumont/soupy:soupy.py:Node.find_next_sibling#11948"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ChrisBeaumont/soupy/blob/795f2f61f711f574d5218fc8a3375d02bda1104f/soupy.py#L1018-L1023", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ChrisBeaumont-soupy:Node.find_next_sibling-f11143d934", "question": "Like :meth:`find`, but searches through :attr:`next_siblings`", "scope": {"holder": "ChrisBeaumont-soupy", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:CiscoDevNet/webexteamssdk:webexteamssdk/api/access_tokens.py:AccessTokensAPI.get#8285", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "webexteamssdk/api/access_tokens.py", "symbol": "AccessTokensAPI.get"}, "source_id": "csn:CiscoDevNet/webexteamssdk:webexteamssdk/api/access_tokens.py:AccessTokensAPI.get#8285"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/CiscoDevNet/webexteamssdk/blob/6fc2cc3557e080ba4b2a380664cb2a0532ae45cd/webexteamssdk/api/access_tokens.py#L104-L148", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:CiscoDevNet-webexteamssdk:AccessTokensAPI.get-267a9746b6", "question": "Exchange an Authorization Code for an Access Token.\n\n Exchange an Authorization Code for an Access Token that can be used to\n invoke the APIs.\n\n Args:\n client_id(basestring): Provided when you created your integration.\n client_secret(basestring): Provided when you created your\n integration.\n code(basestring): The Authorization Code provided by the user\n OAuth process.\n redirect_uri(basestring): The redirect URI used in the user OAuth\n process.\n\n Returns:\n AccessToken: An AccessToken object with the access token provided\n by the Webex Teams cloud.\n\n Raises:\n TypeError: If the parameter types are incorrect.\n ApiError: If the Webex Teams cloud returns an error.", "scope": {"holder": "CiscoDevNet-webexteamssdk", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:CiscoDevNet/webexteamssdk:webexteamssdk/api/team_memberships.py:TeamMembershipsAPI.delete#8230", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "webexteamssdk/api/team_memberships.py", "symbol": "TeamMembershipsAPI.delete"}, "source_id": "csn:CiscoDevNet/webexteamssdk:webexteamssdk/api/team_memberships.py:TeamMembershipsAPI.delete#8230"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/CiscoDevNet/webexteamssdk/blob/6fc2cc3557e080ba4b2a380664cb2a0532ae45cd/webexteamssdk/api/team_memberships.py#L221-L235", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:CiscoDevNet-webexteamssdk:TeamMembershipsAPI.delete-0f9de9b3ca", "question": "Delete a team membership, by ID.\n\n Args:\n membershipId(basestring): The team membership ID.\n\n Raises:\n TypeError: If the parameter types are incorrect.\n ApiError: If the Webex Teams cloud returns an error.", "scope": {"holder": "CiscoDevNet-webexteamssdk", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Clinical-Genomics/scout:scout/adapter/mongo/acmg.py:ACMGHandler.get_evaluations#11657", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "scout/adapter/mongo/acmg.py", "symbol": "ACMGHandler.get_evaluations"}, "source_id": "csn:Clinical-Genomics/scout:scout/adapter/mongo/acmg.py:ACMGHandler.get_evaluations#11657"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Clinical-Genomics/scout/blob/90a551e2e1653a319e654c2405c2866f93d0ebb9/scout/adapter/mongo/acmg.py#L89-L100", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Clinical-Genomics-scout:ACMGHandler.get_evaluations-1c00a0693a", "question": "Return all evaluations for a certain variant.\n\n Args:\n variant_obj (dict): variant dict from the database\n\n Returns:\n pymongo.cursor: database cursor", "scope": {"holder": "Clinical-Genomics-scout", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Clinical-Genomics/scout:scout/parse/matchmaker.py:genomic_features#11364", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "scout/parse/matchmaker.py", "symbol": "genomic_features"}, "source_id": "csn:Clinical-Genomics/scout:scout/parse/matchmaker.py:genomic_features#11364"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Clinical-Genomics/scout/blob/90a551e2e1653a319e654c2405c2866f93d0ebb9/scout/parse/matchmaker.py#L59-L139", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Clinical-Genomics-scout:genomic_features-3f938397ee", "question": "Extract and parse matchmaker-like genomic features from pinned variants\n of a patient\n Args:\n store(MongoAdapter) : connection to the database\n case_obj(dict): a scout case object\n sample_name(str): sample display name\n genes_only(bool): if True only gene names will be included in genomic features\n\n Returns:\n g_features(list): a list of genomic feature objects that looks like this:\n [\n {\n \"gene\": {\n \"id\": \"LIMS2\"\n },\n \"variant\": {\n \"alternateBases\": \"C\",\n \"assembly\": \"GRCh37\",\n \"end\": 128412081,\n \"referenceBases\": \"G\",\n \"referenceName\": \"2\",\n \"start\": 128412080\n },\n \"zygosity\": 1\n },\n ....\n ]", "scope": {"holder": "Clinical-Genomics-scout", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ContinuumIO/pycosat:examples/sudoku.py:solve#10437", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "examples/sudoku.py", "symbol": "solve"}, "source_id": "csn:ContinuumIO/pycosat:examples/sudoku.py:solve#10437"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ContinuumIO/pycosat/blob/b38fd85b6f4dcc18efd6027e96e5785104f53bb0/examples/sudoku.py#L64-L91", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ContinuumIO-pycosat:solve-45aaf4c893", "question": "solve a Sudoku grid inplace", "scope": {"holder": "ContinuumIO-pycosat", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ContinuumIO/pycosat:examples/sudoku.py:sudoku_clauses#10436", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "examples/sudoku.py", "symbol": "sudoku_clauses"}, "source_id": "csn:ContinuumIO/pycosat:examples/sudoku.py:sudoku_clauses#10436"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ContinuumIO/pycosat/blob/b38fd85b6f4dcc18efd6027e96e5785104f53bb0/examples/sudoku.py#L24-L61", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ContinuumIO-pycosat:sudoku_clauses-9b02fbecc9", "question": "Create the (11745) Sudoku clauses, and return them as a list.\n Note that these clauses are *independent* of the particular\n Sudoku puzzle at hand.", "scope": {"holder": "ContinuumIO-pycosat", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Crypto-toolbox/btfxwss:btfxwss/client.py:BtfxWss.authenticate#8409", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "btfxwss/client.py", "symbol": "BtfxWss.authenticate"}, "source_id": "csn:Crypto-toolbox/btfxwss:btfxwss/client.py:BtfxWss.authenticate#8409"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Crypto-toolbox/btfxwss/blob/16827fa6aacb2c0e289aa852bf61a18df6905835/btfxwss/client.py#L326-L334", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Crypto-toolbox-btfxwss:BtfxWss.authenticate-b0dd13b34c", "question": "Authenticate with the Bitfinex API.\n\n :return:", "scope": {"holder": "Crypto-toolbox-btfxwss", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Crypto-toolbox/btfxwss:btfxwss/queue_processor.py:QueueProcessor._handle_trades#8394", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "btfxwss/queue_processor.py", "symbol": "QueueProcessor._handle_trades"}, "source_id": "csn:Crypto-toolbox/btfxwss:btfxwss/queue_processor.py:QueueProcessor._handle_trades#8394"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Crypto-toolbox/btfxwss/blob/16827fa6aacb2c0e289aa852bf61a18df6905835/btfxwss/queue_processor.py#L310-L322", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Crypto-toolbox-btfxwss:QueueProcessor._handle_trades-226d7ce83a", "question": "Files trades in self._trades[chan_id].\n\n :param dtype:\n :param data:\n :param ts:\n :return:", "scope": {"holder": "Crypto-toolbox-btfxwss", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Danielhiversen/PyXiaomiGateway:xiaomi_gateway/__init__.py:XiaomiGateway.get_from_hub#5386", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "xiaomi_gateway/__init__.py", "symbol": "XiaomiGateway.get_from_hub"}, "source_id": "csn:Danielhiversen/PyXiaomiGateway:xiaomi_gateway/__init__.py:XiaomiGateway.get_from_hub#5386"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Danielhiversen/PyXiaomiGateway/blob/21b38ab972d67402f2124dba02101ddfd8d9e0b4/xiaomi_gateway/__init__.py#L379-L384", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Danielhiversen-PyXiaomiGateway:XiaomiGateway.get_from_hub-bf88a27e71", "question": "Get data from gateway", "scope": {"holder": "Danielhiversen-PyXiaomiGateway", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Danielhiversen/PyXiaomiGateway:xiaomi_gateway/__init__.py:XiaomiGateway.write_to_hub#5385", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "xiaomi_gateway/__init__.py", "symbol": "XiaomiGateway.write_to_hub"}, "source_id": "csn:Danielhiversen/PyXiaomiGateway:xiaomi_gateway/__init__.py:XiaomiGateway.write_to_hub#5385"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Danielhiversen/PyXiaomiGateway/blob/21b38ab972d67402f2124dba02101ddfd8d9e0b4/xiaomi_gateway/__init__.py#L330-L377", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Danielhiversen-PyXiaomiGateway:XiaomiGateway.write_to_hub-0749ded7c5", "question": "Send data to gateway to turn on / off device", "scope": {"holder": "Danielhiversen-PyXiaomiGateway", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Dirrot/python-cryptocoincharts-api:CryptoCoinChartsApi/CryptoCoinChartsApi.py:API._getdata#19261", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "CryptoCoinChartsApi/CryptoCoinChartsApi.py", "symbol": "API._getdata"}, "source_id": "csn:Dirrot/python-cryptocoincharts-api:CryptoCoinChartsApi/CryptoCoinChartsApi.py:API._getdata#19261"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Dirrot/python-cryptocoincharts-api/blob/8bf7a35c1032847aaea322b304014cd52853c273/CryptoCoinChartsApi/CryptoCoinChartsApi.py#L98-L117", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Dirrot-python-cryptocoincharts-api:API._getdata-0d0af356c7", "question": "Wrapper method", "scope": {"holder": "Dirrot-python-cryptocoincharts-api", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Dirrot/python-cryptocoincharts-api:CryptoCoinChartsApi/CryptoCoinChartsApi.py:API.tradingpair#19259", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "CryptoCoinChartsApi/CryptoCoinChartsApi.py", "symbol": "API.tradingpair"}, "source_id": "csn:Dirrot/python-cryptocoincharts-api:CryptoCoinChartsApi/CryptoCoinChartsApi.py:API.tradingpair#19259"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Dirrot/python-cryptocoincharts-api/blob/8bf7a35c1032847aaea322b304014cd52853c273/CryptoCoinChartsApi/CryptoCoinChartsApi.py#L44-L65", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Dirrot-python-cryptocoincharts-api:API.tradingpair-d8ea6d052c", "question": "Use this function to query price and volume data for ONE trading pair.\n A list with all coin currencies can be found by using the listcoins method.\n A example pair: currency1_currency2 = \"doge_btc\" \n Usage: http://api.cryptocoincharts.info/tradingPair/[currency1_currency2]", "scope": {"holder": "Dirrot-python-cryptocoincharts-api", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:DiscordBotList/DBL-Python-Library:dbl/client.py:Client.close#7432", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "dbl/client.py", "symbol": "Client.close"}, "source_id": "csn:DiscordBotList/DBL-Python-Library:dbl/client.py:Client.close#7432"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/DiscordBotList/DBL-Python-Library/blob/c1461ae0acc644cdeedef8fd6b5e36f76d81c1aa/dbl/client.py#L351-L359", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:DiscordBotList-DBL-Python-Library:Client.close-80f7d55540", "question": "This function is a coroutine.\n\n Closes all connections.", "scope": {"holder": "DiscordBotList-DBL-Python-Library", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:DiscordBotList/DBL-Python-Library:dbl/client.py:Client.get_widget_small#7431", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "dbl/client.py", "symbol": "Client.get_widget_small"}, "source_id": "csn:DiscordBotList/DBL-Python-Library:dbl/client.py:Client.get_widget_small#7431"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/DiscordBotList/DBL-Python-Library/blob/c1461ae0acc644cdeedef8fd6b5e36f76d81c1aa/dbl/client.py#L330-L349", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:DiscordBotList-DBL-Python-Library:Client.get_widget_small-9b8ab5ab63", "question": "This function is a coroutine.\n\n Generates the default small widget.\n\n Parameters\n ==========\n\n bot_id: int\n The bot_id of the bot you wish to make a widget for.\n\n Returns\n =======\n\n URL of the widget: str", "scope": {"holder": "DiscordBotList-DBL-Python-Library", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:EpistasisLab/scikit-mdr:mdr/continuous_mdr.py:ContinuousMDR.fit#13176", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "mdr/continuous_mdr.py", "symbol": "ContinuousMDR.fit"}, "source_id": "csn:EpistasisLab/scikit-mdr:mdr/continuous_mdr.py:ContinuousMDR.fit#13176"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/EpistasisLab/scikit-mdr/blob/768565deb10467d04a960d27e000ab38b7aa8a62/mdr/continuous_mdr.py#L57-L93", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:EpistasisLab-scikit-mdr:ContinuousMDR.fit-c13b97b5d4", "question": "Constructs the Continuous MDR feature map from the provided training data.\n\n Parameters\n ----------\n features: array-like {n_samples, n_features}\n Feature matrix\n targets: array-like {n_samples}\n List of target values for prediction\n\n Returns\n -------\n self: A copy of the fitted model", "scope": {"holder": "EpistasisLab-scikit-mdr", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:EpistasisLab/scikit-mdr:mdr/utils/utils.py:conditional_entropy#13182", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "mdr/utils/utils.py", "symbol": "conditional_entropy"}, "source_id": "csn:EpistasisLab/scikit-mdr:mdr/utils/utils.py:conditional_entropy#13182"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/EpistasisLab/scikit-mdr/blob/768565deb10467d04a960d27e000ab38b7aa8a62/mdr/utils/utils.py#L68-L86", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:EpistasisLab-scikit-mdr:conditional_entropy-63aec33569", "question": "Calculates the conditional entropy, H(X|Y), in the given base\n\n Parameters\n ----------\n X: array-like (# samples)\n An array of values for which to compute the conditional entropy\n Y: array-like (# samples)\n An array of values for which to compute the conditional entropy\n base: integer (default: 2)\n The base in which to calculate conditional entropy\n\n Returns\n ----------\n conditional_entropy: float\n The conditional entropy calculated according to the equation H(X|Y) = H(X,Y) - H(Y)", "scope": {"holder": "EpistasisLab-scikit-mdr", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Erotemic/progiter:setup.py:parse_description#12079", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "setup.py", "symbol": "parse_description"}, "source_id": "csn:Erotemic/progiter:setup.py:parse_description#12079"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Erotemic/progiter/blob/24f1ad15d79f76cccef7b5811d341ab33b72bf1e/setup.py#L71-L104", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Erotemic-progiter:parse_description-3fccbb0f21", "question": "Parse the description in the README file\n\n CommandLine:\n python -c \"import setup; print(setup.parse_description())\"", "scope": {"holder": "Erotemic-progiter", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Erotemic/xdoctest:make_rtd.py:modify_conf#12249", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "make_rtd.py", "symbol": "modify_conf"}, "source_id": "csn:Erotemic/xdoctest:make_rtd.py:modify_conf#12249"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Erotemic/xdoctest/blob/1c85209498bcf3791c919985b3d1904ec2339768/make_rtd.py#L42-L65", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Erotemic-xdoctest:modify_conf-3cdbdd2a0e", "question": "pip install redbaron", "scope": {"holder": "Erotemic-xdoctest", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:GGiecold/Density_Sampling:Density_Sampling.py:density_sampling#15575", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "Density_Sampling.py", "symbol": "density_sampling"}, "source_id": "csn:GGiecold/Density_Sampling:Density_Sampling.py:density_sampling#15575"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/GGiecold/Density_Sampling/blob/8c8e6c63a97fecf958238e12947e5e6542b64102/Density_Sampling.py#L212-L323", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:GGiecold-Density_Sampling:density_sampling-d2c268f1ca", "question": "The i-th sample point of the data-set 'data' is selected by density sampling\n with a probability given by:\n \n | 0 if outlier_density > LD[i];\n P(keep the i-th data-point) = | 1 if outlier_density <= LD[i] <= target_density;\n | target_density / LD[i] if LD[i] > target_density.\n \n Here 'LD[i]' denotes the local density of the i-th sample point of the data-set, \n whereas 'outlier_density' and 'target_density' are computed as particular percentiles\n of that distribution of local densities.\n \n Parameters\n ----------\n data : array of shape (n_samples, n_features)\n The data-set, a fraction of whose sample points will be extracted\n by density sampling.\n \n local_densities : array of shape (n_samples,), optional (default = None)\n The i-th entry of this vector corresponds to the local density of the i-th sample\n point in the order of the rows of 'data'.\n \n metric : string, optional (default = 'manhattan')\n The distance metric used to determine the nearest-neighbor to each data-point.\n The DistanceMetric class defined in scikit-learn's library lists all available\n metrics.\n \n kernel_mult : float, optional (default = 2.0)\n The kernel multiplier, which determine (in terms of the median of the distribution\n of distances among nearest neighbors) the extent of the regions centered\n around each sample point to consider for the computation of the local density\n associated to that particular sample point.\n \n outlier_percentile : float, optional (default = 0.01)\n Specify the outlier density as a percentile of the distribution of local densities.\n \n target_percentile : float, optional (default = 0.05)\n Specifiy the target density as a percentile of the distribution of local densities.\n Relevant only if 'desired_samples' is left unspecified. \n \n desired_samples : int, optional (default = None)\n The number of samples to be selected from the whole data-set such that members\n of rare populations and members of more common populations are roughly\n equally represented. To that purpose, a target density is computed that to selects about\n 'desired_samples' data-points.\n \n Returns\n -------\n samples_kept : array of shape (n_selected_samples,)\n If the 'i'-th sample point of 'data' has been selected by a given instance of\n density sampling, number 'i' is featured in the array returned by \n the present function.", "scope": {"holder": "GGiecold-Density_Sampling", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:GGiecold/Density_Sampling:Density_Sampling.py:get_local_densities#15574", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "Density_Sampling.py", "symbol": "get_local_densities"}, "source_id": "csn:GGiecold/Density_Sampling:Density_Sampling.py:get_local_densities#15574"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/GGiecold/Density_Sampling/blob/8c8e6c63a97fecf958238e12947e5e6542b64102/Density_Sampling.py#L147-L209", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:GGiecold-Density_Sampling:get_local_densities-e1acf061bb", "question": "For each sample point of the data-set 'data', estimate a local density in feature\n space by counting the number of neighboring data-points within a particular\n region centered around that sample point.\n \n Parameters\n ----------\n data : array of shape (n_samples, n_features)\n The data-set, a fraction of whose sample points will be extracted\n by density sampling.\n \n kernel_mult : float, optional (default = 2.0)\n The kernel multiplier, which determine (in terms of the median of the distribution\n of distances among nearest neighbors) the extent of the regions centered\n around each sample point to consider for the computation of the local density\n associated to that particular sample point.\n \n metric : string, optional (default = 'manhattan')\n The distance metric used to determine the nearest-neighbor to each data-point.\n The DistanceMetric class defined in scikit-learn's library lists all available\n metrics.\n \n Returns\n -------\n local_densities : array of shape (n_samples,)\n The i-th entry of this vector corresponds to the local density of the i-th sample\n point in the order of the rows of 'data'.", "scope": {"holder": "GGiecold-Density_Sampling", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Gandi/pyramid_kvs:pyramid_kvs/perlsess.py:PerlSession.connect#13744", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyramid_kvs/perlsess.py", "symbol": "PerlSession.connect"}, "source_id": "csn:Gandi/pyramid_kvs:pyramid_kvs/perlsess.py:PerlSession.connect#13744"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Gandi/pyramid_kvs/blob/36285f2e50d8181428f383f6fc1d79a34ea9ac3c/pyramid_kvs/perlsess.py#L31-L38", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Gandi-pyramid_kvs:PerlSession.connect-df42aed15c", "question": "Call that method in the pyramid configuration phase.", "scope": {"holder": "Gandi-pyramid_kvs", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:GeoffAtHome/lightwave:lightwave/lightwave.py:LWLink._send_message#19266", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "lightwave/lightwave.py", "symbol": "LWLink._send_message"}, "source_id": "csn:GeoffAtHome/lightwave:lightwave/lightwave.py:LWLink._send_message#19266"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/GeoffAtHome/lightwave/blob/2fab4ee8c9f14dd97dffd4b8cd70b217e884e581/lightwave/lightwave.py#L29-L34", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:GeoffAtHome-lightwave:LWLink._send_message-3c92ed8d66", "question": "Add message to queue and start processing the queue.", "scope": {"holder": "GeoffAtHome-lightwave", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:GeoffAtHome/lightwave:lightwave/lightwave.py:LWLink.turn_on_light#19267", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "lightwave/lightwave.py", "symbol": "LWLink.turn_on_light"}, "source_id": "csn:GeoffAtHome/lightwave:lightwave/lightwave.py:LWLink.turn_on_light#19267"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/GeoffAtHome/lightwave/blob/2fab4ee8c9f14dd97dffd4b8cd70b217e884e581/lightwave/lightwave.py#L46-L49", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:GeoffAtHome-lightwave:LWLink.turn_on_light-eaf8e4ffc3", "question": "Create the message to turn light on.", "scope": {"holder": "GeoffAtHome-lightwave", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:GreatFruitOmsk/tailhead:tailhead/__init__.py:Tailer.follow#14990", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "tailhead/__init__.py", "symbol": "Tailer.follow"}, "source_id": "csn:GreatFruitOmsk/tailhead:tailhead/__init__.py:Tailer.follow#14990"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/GreatFruitOmsk/tailhead/blob/a3b1324a39935f8ffcfda59328a9a458672889d9/tailhead/__init__.py#L249-L284", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:GreatFruitOmsk-tailhead:Tailer.follow-ffd220f10c", "question": "Iterator generator that returns lines as data is added to the file.\n\n None will be yielded if no new line is available.\n Caller may either wait and re-try or end iteration.", "scope": {"holder": "GreatFruitOmsk-tailhead", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:GreatFruitOmsk/tailhead:tailhead/__init__.py:Tailer.seek_previous_line#14987", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "tailhead/__init__.py", "symbol": "Tailer.seek_previous_line"}, "source_id": "csn:GreatFruitOmsk/tailhead:tailhead/__init__.py:Tailer.seek_previous_line#14987"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/GreatFruitOmsk/tailhead/blob/a3b1324a39935f8ffcfda59328a9a458672889d9/tailhead/__init__.py#L143-L196", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:GreatFruitOmsk-tailhead:Tailer.seek_previous_line-ed714d7b5b", "question": "Seek previous line relative to the current file position.\n\n :return: Position of the line or -1 if previous line was not found.", "scope": {"holder": "GreatFruitOmsk-tailhead", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:HaaLeo/pylint-file-header:pylintfileheader/file_header_checker.py:FileHeaderChecker.process_module#19460", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pylintfileheader/file_header_checker.py", "symbol": "FileHeaderChecker.process_module"}, "source_id": "csn:HaaLeo/pylint-file-header:pylintfileheader/file_header_checker.py:FileHeaderChecker.process_module#19460"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/HaaLeo/pylint-file-header/blob/245421616fdb6b9aca234fd58c503f9a937749f6/pylintfileheader/file_header_checker.py#L41-L61", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:HaaLeo-pylint-file-header:FileHeaderChecker.process_module-d650fe2e6e", "question": "Process the astroid node stream.", "scope": {"holder": "HaaLeo-pylint-file-header", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:HumanBrainProject/hbp-service-client:hbp_service_client/storage_service/api.py:ApiClient._prep_params#15578", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "hbp_service_client/storage_service/api.py", "symbol": "ApiClient._prep_params"}, "source_id": "csn:HumanBrainProject/hbp-service-client:hbp_service_client/storage_service/api.py:ApiClient._prep_params#15578"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/HumanBrainProject/hbp-service-client/blob/b338fb41a7f0e7b9d654ff28fcf13a56d03bff4d/hbp_service_client/storage_service/api.py#L85-L88", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:HumanBrainProject-hbp-service-client:ApiClient._prep_params-4eed5e74df", "question": "Remove empty (None) valued keywords and self from function parameters", "scope": {"holder": "HumanBrainProject-hbp-service-client", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:HumanBrainProject/hbp-service-client:hbp_service_client/storage_service/api.py:ApiClient.list_folder_content#15591", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "hbp_service_client/storage_service/api.py", "symbol": "ApiClient.list_folder_content"}, "source_id": "csn:HumanBrainProject/hbp-service-client:hbp_service_client/storage_service/api.py:ApiClient.list_folder_content#15591"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/HumanBrainProject/hbp-service-client/blob/b338fb41a7f0e7b9d654ff28fcf13a56d03bff4d/hbp_service_client/storage_service/api.py#L659-L715", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:HumanBrainProject-hbp-service-client:ApiClient.list_folder_content-6ca9c1b69a", "question": "List files and folders (not recursively) contained in the folder.\n\n This function does not retrieve all results, pages have\n to be manually retrieved by the caller.\n\n Args:\n folder (str): The UUID of the requested folder.\n name (str): Optional filter on entity name.\n entity_type (str): Optional filter on entity type.\n Admitted values: ['file', 'folder'].\n content_type (str): Optional filter on entity content type (only\n files are returned).\n page_size (int): Number of elements per page.\n page (int): Number of the page.\n ordering (str): Indicate on which fields to sort the result. Prepend\n '-' to invert order. Multiple values can be provided.\n Ordering is supported on: ['name', 'created_on', 'modified_on'].\n Example: 'ordering=name,created_on'\n\n Returns:\n A dictionary of the results::\n\n {\n u'count': 1,\n u'next': None,\n u'previous': None,\n u'results': [{u'content_type': u'plain/text',\n u'created_by': u'303447',\n u'created_on': u'2017-03-13T10:17:01.688472Z',\n u'description': u'',\n u'entity_type': u'file',\n u'modified_by': u'303447',\n u'modified_on': u'2017-03-13T10:17:01.688632Z',\n u'name': u'file_1',\n u'parent': u'eac11058-4ae0-4ea9-ada8-d3ea23887509',\n u'uuid': u'0e17eaac-cb00-4336-b9d7-657026844281'}]\n }\n\n Raises:\n StorageArgumentException: Invalid arguments\n StorageForbiddenException: Server response code 403\n StorageNotFoundException: Server response code 404\n StorageException: other 400-600 error codes", "scope": {"holder": "HumanBrainProject-hbp-service-client", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:HydraChain/hydrachain:hydrachain/consensus/base.py:BlockProposal.validate_votes#16670", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "hydrachain/consensus/base.py", "symbol": "BlockProposal.validate_votes"}, "source_id": "csn:HydraChain/hydrachain:hydrachain/consensus/base.py:BlockProposal.validate_votes#16670"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/HydraChain/hydrachain/blob/6c0919b0575dc8aa481f3a8c703e1a7f0575ecc3/hydrachain/consensus/base.py#L497-L511", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:HydraChain-hydrachain:BlockProposal.validate_votes-ab82735b90", "question": "set of validators may change between heights", "scope": {"holder": "HydraChain-hydrachain", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:HydraChain/hydrachain:hydrachain/consensus/manager.py:RoundManager.get_timeout#16679", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "hydrachain/consensus/manager.py", "symbol": "RoundManager.get_timeout"}, "source_id": "csn:HydraChain/hydrachain:hydrachain/consensus/manager.py:RoundManager.get_timeout#16679"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/HydraChain/hydrachain/blob/6c0919b0575dc8aa481f3a8c703e1a7f0575ecc3/hydrachain/consensus/manager.py#L572-L581", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:HydraChain-hydrachain:RoundManager.get_timeout-8b467feaab", "question": "setup a timeout for waiting for a proposal", "scope": {"holder": "HydraChain-hydrachain", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:IEMLdev/ieml:ieml/dictionary/version.py:DictionaryVersion.load#10376", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "ieml/dictionary/version.py", "symbol": "DictionaryVersion.load"}, "source_id": "csn:IEMLdev/ieml:ieml/dictionary/version.py:DictionaryVersion.load#10376"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/IEMLdev/ieml/blob/4c842ba7e6165e2f1b4a4e2e98759f9f33af5f25/ieml/dictionary/version.py#L145-L164", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:IEMLdev-ieml:DictionaryVersion.load-45e75bb446", "question": "Download the dictionary version and cache the retrieved file.\n :return: None", "scope": {"holder": "IEMLdev-ieml", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:IEMLdev/ieml:scripts/dictionary_tools.py:_add_mode_t#10380", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "scripts/dictionary_tools.py", "symbol": "_add_mode_t"}, "source_id": "csn:IEMLdev/ieml:scripts/dictionary_tools.py:_add_mode_t#10380"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/IEMLdev/ieml/blob/4c842ba7e6165e2f1b4a4e2e98759f9f33af5f25/scripts/dictionary_tools.py#L106-L112", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:IEMLdev-ieml:_add_mode_t-ebbf7708da", "question": "O:O:.O:O:.- => O:O:.O:O:.t.-", "scope": {"holder": "IEMLdev-ieml", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:IdentityPython/fedoidcmsg:src/fedoidcmsg/entity.py:FederationEntity.pick_signed_metadata_statements#19665", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/fedoidcmsg/entity.py", "symbol": "FederationEntity.pick_signed_metadata_statements"}, "source_id": "csn:IdentityPython/fedoidcmsg:src/fedoidcmsg/entity.py:FederationEntity.pick_signed_metadata_statements#19665"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/IdentityPython/fedoidcmsg/blob/d30107be02521fa6cdfe285da3b6b0cdd153c8cc/src/fedoidcmsg/entity.py#L91-L105", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:IdentityPython-fedoidcmsg:FederationEntity.pick_signed_metadata_statements-6c7c2b59b5", "question": "Pick signed metadata statements based on ISS pattern matching\n \n :param fo: Federation operators ID\n :param context: In connect with which operation (one of the values in \n :py:data:`fedoidc.CONTEXTS`). \n :return: list of tuples (FO ID, signed metadata statement)", "scope": {"holder": "IdentityPython-fedoidcmsg", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:IdentityPython/fedoidcmsg:src/fedoidcmsg/operator.py:LessOrEqual.eval#19547", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/fedoidcmsg/operator.py", "symbol": "LessOrEqual.eval"}, "source_id": "csn:IdentityPython/fedoidcmsg:src/fedoidcmsg/operator.py:LessOrEqual.eval#19547"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/IdentityPython/fedoidcmsg/blob/d30107be02521fa6cdfe285da3b6b0cdd153c8cc/src/fedoidcmsg/operator.py#L88-L117", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:IdentityPython-fedoidcmsg:LessOrEqual.eval-10eb3f18aa", "question": "Apply the less or equal algorithm on the ordered list of metadata\n statements\n \n :param orig: Start values\n :return:", "scope": {"holder": "IdentityPython-fedoidcmsg", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:JIC-CSB/jicbioimage.illustrate:jicbioimage/illustrate/__init__.py:AnnotatedImage.from_grayscale#22167", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "jicbioimage/illustrate/__init__.py", "symbol": "AnnotatedImage.from_grayscale"}, "source_id": "csn:JIC-CSB/jicbioimage.illustrate:jicbioimage/illustrate/__init__.py:AnnotatedImage.from_grayscale#22167"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/JIC-CSB/jicbioimage.illustrate/blob/d88ddf81ee3eb3949677e2ef746af8169ce88092/jicbioimage/illustrate/__init__.py#L159-L171", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:JIC-CSB-jicbioimage.illustrate:AnnotatedImage.from_grayscale-b20bbf803e", "question": "Return a canvas from a grayscale image.\n\n :param im: single channel image\n :channels_on: channels to populate with input image\n :returns: :class:`jicbioimage.illustrate.Canvas`", "scope": {"holder": "JIC-CSB-jicbioimage.illustrate", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:JIC-CSB/jicbioimage.illustrate:jicbioimage/illustrate/__init__.py:Canvas.draw_cross#22164", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "jicbioimage/illustrate/__init__.py", "symbol": "Canvas.draw_cross"}, "source_id": "csn:JIC-CSB/jicbioimage.illustrate:jicbioimage/illustrate/__init__.py:Canvas.draw_cross#22164"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/JIC-CSB/jicbioimage.illustrate/blob/d88ddf81ee3eb3949677e2ef746af8169ce88092/jicbioimage/illustrate/__init__.py#L65-L86", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:JIC-CSB-jicbioimage.illustrate:Canvas.draw_cross-7a817e634e", "question": "Draw a cross on the canvas.\n\n :param position: (row, col) tuple\n :param color: RGB tuple\n :param radius: radius of the cross (int)", "scope": {"holder": "JIC-CSB-jicbioimage.illustrate", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Jasily/jasily-python:jasily/collection/funcs.py:CallableList.insert_func#21432", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "jasily/collection/funcs.py", "symbol": "CallableList.insert_func"}, "source_id": "csn:Jasily/jasily-python:jasily/collection/funcs.py:CallableList.insert_func#21432"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Jasily/jasily-python/blob/1c821a120ebbbbc3c5761f5f1e8a73588059242a/jasily/collection/funcs.py#L48-L53", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Jasily-jasily-python:CallableList.insert_func-35c52d094b", "question": "insert func with given arguments and keywords.", "scope": {"holder": "Jasily-jasily-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Jasily/jasily-python:jasily/utils/__init__.py:jrepr#21314", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "jasily/utils/__init__.py", "symbol": "jrepr"}, "source_id": "csn:Jasily/jasily-python:jasily/utils/__init__.py:jrepr#21314"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Jasily/jasily-python/blob/1c821a120ebbbbc3c5761f5f1e8a73588059242a/jasily/utils/__init__.py#L11-L18", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Jasily-jasily-python:jrepr-605526a928", "question": "customized `repr()`.", "scope": {"holder": "Jasily-jasily-python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Jaymon/pyt:pyt/path.py:PathFinder._find_basename#16119", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyt/path.py", "symbol": "PathFinder._find_basename"}, "source_id": "csn:Jaymon/pyt:pyt/path.py:PathFinder._find_basename#16119"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Jaymon/pyt/blob/801581fd0ae238158134bde1c937fa199fa626b2/pyt/path.py#L286-L361", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Jaymon-pyt:PathFinder._find_basename-fea83cb441", "question": "check if name combined with test prefixes or postfixes is found anywhere\n in the list of basenames\n\n :param name: string, the name you're searching for\n :param basenames: list, a list of basenames to check\n :param is_prefix: bool, True if this is a prefix search, which means it will\n also check if name matches any of the basenames without the prefixes or\n postfixes, if it is False then the prefixes or postfixes must be present\n (ie, the module we're looking for is the actual test module, not the parent\n modules it's contained in)\n :returns: string, the basename if it is found", "scope": {"holder": "Jaymon-pyt", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Jaymon/pyt:pyt/path.py:PathFinder.paths#16123", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyt/path.py", "symbol": "PathFinder.paths"}, "source_id": "csn:Jaymon/pyt:pyt/path.py:PathFinder.paths#16123"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Jaymon/pyt/blob/801581fd0ae238158134bde1c937fa199fa626b2/pyt/path.py#L490-L559", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Jaymon-pyt:PathFinder.paths-d44bab332e", "question": "given a basedir, yield all test modules paths recursively found in\n basedir that are test modules\n\n return -- generator", "scope": {"holder": "Jaymon-pyt", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Jaza/s3-saver:s3_saver.py:S3Saver._delete_local#18501", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "s3_saver.py", "symbol": "S3Saver._delete_local"}, "source_id": "csn:Jaza/s3-saver:s3_saver.py:S3Saver._delete_local#18501"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Jaza/s3-saver/blob/81dc4447d76c2fc0b0238fb96fa70e879612e355/s3_saver.py#L53-L57", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Jaza-s3-saver:S3Saver._delete_local-e76c96ca33", "question": "Deletes the specified file from the local filesystem.", "scope": {"holder": "Jaza-s3-saver", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Jaza/s3-saver:s3_saver.py:S3Saver.delete#18503", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "s3_saver.py", "symbol": "S3Saver.delete"}, "source_id": "csn:Jaza/s3-saver:s3_saver.py:S3Saver.delete#18503"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Jaza/s3-saver/blob/81dc4447d76c2fc0b0238fb96fa70e879612e355/s3_saver.py#L77-L86", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Jaza-s3-saver:S3Saver.delete-bf177bed80", "question": "Deletes the specified file, either locally or from S3, depending on the file's storage type.", "scope": {"holder": "Jaza-s3-saver", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:JonathanRaiman/ciseau:ciseau/sentence_tokenizer.py:detect_sentence_boundaries#14592", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "ciseau/sentence_tokenizer.py", "symbol": "detect_sentence_boundaries"}, "source_id": "csn:JonathanRaiman/ciseau:ciseau/sentence_tokenizer.py:detect_sentence_boundaries#14592"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/JonathanRaiman/ciseau/blob/f72d1c82d85eeb3d3ac9fac17690041725402175/ciseau/sentence_tokenizer.py#L15-L96", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:JonathanRaiman-ciseau:detect_sentence_boundaries-1451a4cd9e", "question": "Subdivide an input list of strings (tokens)\n into multiple lists according to detected\n sentence boundaries.\n\n ```\n detect_sentence_boundaries(\n [\"Cat \", \"sat \", \"mat\", \". \", \"Cat \", \"'s \", \"named \", \"Cool\", \".\"]\n )\n #=> [\n [\"Cat \", \"sat \", \"mat\", \". \"],\n [\"Cat \", \"'s \", \"named \", \"Cool\", \".\"]\n ]\n ```\n\n Arguments:\n ----------\n\n tokens : list\n\n Returns:\n --------\n list> : original list subdivided into multiple\n lists according to (detected) sentence boundaries.", "scope": {"holder": "JonathanRaiman-ciseau", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:JonathanRaiman/ciseau:ciseau/word_tokenizer.py:split_with_locations#14597", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "ciseau/word_tokenizer.py", "symbol": "split_with_locations"}, "source_id": "csn:JonathanRaiman/ciseau:ciseau/word_tokenizer.py:split_with_locations#14597"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/JonathanRaiman/ciseau/blob/f72d1c82d85eeb3d3ac9fac17690041725402175/ciseau/word_tokenizer.py#L112-L138", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:JonathanRaiman-ciseau:split_with_locations-6550dbdf39", "question": "Use an integer list to split the string\n contained in `text`.\n\n Arguments:\n ----------\n text : str, same length as locations.\n locations : list, contains values\n 'SHOULD_SPLIT', 'UNDECIDED', and\n 'SHOULD_NOT_SPLIT'. Will create\n strings between each 'SHOULD_SPLIT'\n locations.\n Returns:\n --------\n Generator : the substrings of text\n corresponding to the slices given\n in locations.", "scope": {"holder": "JonathanRaiman-ciseau", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:JukeboxPipeline/jukedj:src/jukedj/models.py:Project.path#19618", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/jukedj/models.py", "symbol": "Project.path"}, "source_id": "csn:JukeboxPipeline/jukedj:src/jukedj/models.py:Project.path#19618"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/JukeboxPipeline/jukedj/blob/d4159961c819c26792a278981ee68106ee15f3f3/src/jukedj/models.py#L88-L98", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:JukeboxPipeline-jukedj:Project.path-1e64360368", "question": "Set path\n\n :param value: The value for path\n :type value: str\n :raises: None", "scope": {"holder": "JukeboxPipeline-jukedj", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:JukeboxPipeline/jukedj:src/jukedj/models.py:add_default_deps#19610", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/jukedj/models.py", "symbol": "add_default_deps"}, "source_id": "csn:JukeboxPipeline/jukedj:src/jukedj/models.py:add_default_deps#19610"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/JukeboxPipeline/jukedj/blob/d4159961c819c26792a278981ee68106ee15f3f3/src/jukedj/models.py#L366-L380", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:JukeboxPipeline-jukedj:add_default_deps-27868f6de1", "question": "Add or create the default departments for the given project\n\n :param project: the project that needs default departments\n :type project: :class:`muke.models.Project`\n :returns: None\n :rtype: None\n :raises: None", "scope": {"holder": "JukeboxPipeline-jukedj", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Julian/Ivoire:ivoire/result.py:DotsFormatter.result_summary#15234", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "ivoire/result.py", "symbol": "DotsFormatter.result_summary"}, "source_id": "csn:Julian/Ivoire:ivoire/result.py:DotsFormatter.result_summary#15234"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Julian/Ivoire/blob/5b8218cffa409ed733cf850a6fde16fafb8fc2af/ivoire/result.py#L206-L214", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Julian-Ivoire:DotsFormatter.result_summary-f060c3d93b", "question": "Return a summary of the results.", "scope": {"holder": "Julian-Ivoire", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Julian/Ivoire:ivoire/transform.py:ExampleTransformer.visit_With#15239", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "ivoire/transform.py", "symbol": "ExampleTransformer.visit_With"}, "source_id": "csn:Julian/Ivoire:ivoire/transform.py:ExampleTransformer.visit_With#15239"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Julian/Ivoire/blob/5b8218cffa409ed733cf850a6fde16fafb8fc2af/ivoire/transform.py#L32-L53", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Julian-Ivoire:ExampleTransformer.visit_With-d31c6a9fe3", "question": "with describe(thing) as it:\n ...\n\n |\n v\n\n class TestThing(TestCase):\n ...", "scope": {"holder": "Julian-Ivoire", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Julian/Virtue:virtue/cli.py:main#18325", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "virtue/cli.py", "symbol": "main"}, "source_id": "csn:Julian/Virtue:virtue/cli.py:main#18325"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Julian/Virtue/blob/d08be37d759c38c94a160bc13fe8f51bb2aeeedd/virtue/cli.py#L61-L70", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Julian-Virtue:main-74acba3b52", "question": "virtue discovers and runs tests found in the given objects.\n\n Provide it with one or more tests (packages, modules or objects) to run.", "scope": {"holder": "Julian-Virtue", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Julian/Virtue:virtue/locators.py:prefixed_by#18305", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "virtue/locators.py", "symbol": "prefixed_by"}, "source_id": "csn:Julian/Virtue:virtue/locators.py:prefixed_by#18305"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Julian/Virtue/blob/d08be37d759c38c94a160bc13fe8f51bb2aeeedd/virtue/locators.py#L11-L25", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Julian-Virtue:prefixed_by-f5ba260646", "question": "Make a callable returning True for names starting with the given prefix.\n\n The returned callable takes two arguments, the attribute or name of\n the object, and possibly its corresponding value (which is ignored),\n as suitable for use with :meth:`ObjectLocator.is_test_module` and\n :meth:`ObjectLocator.is_test_method`\\ .", "scope": {"holder": "Julian-Virtue", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Kroisse/FormEncode-Jinja2:formencode_jinja2/formfill.py:default_formatter#19919", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "formencode_jinja2/formfill.py", "symbol": "default_formatter"}, "source_id": "csn:Kroisse/FormEncode-Jinja2:formencode_jinja2/formfill.py:default_formatter#19919"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Kroisse/FormEncode-Jinja2/blob/5e8c97ec282bd85aa39f9ff83cf31fd27ed7b594/formencode_jinja2/formfill.py#L114-L117", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Kroisse-FormEncode-Jinja2:default_formatter-1cc91aa292", "question": "Escape the error, and wrap it in a span with class ``error-message``", "scope": {"holder": "Kroisse-FormEncode-Jinja2", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:L3viathan/sql-mojo-parser:sql_mojo_parser/__init__.py:p_colspec#20164", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sql_mojo_parser/__init__.py", "symbol": "p_colspec"}, "source_id": "csn:L3viathan/sql-mojo-parser:sql_mojo_parser/__init__.py:p_colspec#20164"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/L3viathan/sql-mojo-parser/blob/fc460c42f3fbcc21c6fc08c0aede8e2a5db637f3/sql_mojo_parser/__init__.py#L103-L125", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:L3viathan-sql-mojo-parser:p_colspec-edf7d0d08c", "question": "colspec : STAR\n | NAME\n | function\n | NAME COMMA colspec\n | function COMMA colspec", "scope": {"holder": "L3viathan-sql-mojo-parser", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:L3viathan/sql-mojo-parser:sql_mojo_parser/__init__.py:t_NAME#20161", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sql_mojo_parser/__init__.py", "symbol": "t_NAME"}, "source_id": "csn:L3viathan/sql-mojo-parser:sql_mojo_parser/__init__.py:t_NAME#20161"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/L3viathan/sql-mojo-parser/blob/fc460c42f3fbcc21c6fc08c0aede8e2a5db637f3/sql_mojo_parser/__init__.py#L19-L26", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:L3viathan-sql-mojo-parser:t_NAME-43e8b2b86a", "question": "r'[A-Za-z_][A-Za-z0-9_]*", "scope": {"holder": "L3viathan-sql-mojo-parser", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:LLNL/scraper:scraper/github/queryManager.py:DataManager.fileLoad#10327", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "scraper/github/queryManager.py", "symbol": "DataManager.fileLoad"}, "source_id": "csn:LLNL/scraper:scraper/github/queryManager.py:DataManager.fileLoad#10327"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LLNL/scraper/blob/881a316e4c04dfa5a9cf491b7c7f9f997a7c56ea/scraper/github/queryManager.py#L454-L478", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LLNL-scraper:DataManager.fileLoad-a7429ae0c1", "question": "Load a JSON data file into the internal JSON data dictionary.\n\n Current internal data will be overwritten.\n If no file path is provided, the stored data file path will be used.\n\n Args:\n filePath (Optional[str]): A relative or absolute path to a\n '.json' file. Defaults to None.\n updatePath (Optional[bool]): Specifies whether or not to update\n the stored data file path. Defaults to True.", "scope": {"holder": "LLNL-scraper", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LLNL/scraper:scripts/get_users_emails.py:GitHub_Users_Emails.write_to_file#10313", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "scripts/get_users_emails.py", "symbol": "GitHub_Users_Emails.write_to_file"}, "source_id": "csn:LLNL/scraper:scripts/get_users_emails.py:GitHub_Users_Emails.write_to_file#10313"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LLNL/scraper/blob/881a316e4c04dfa5a9cf491b7c7f9f997a7c56ea/scripts/get_users_emails.py#L104-L114", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LLNL-scraper:GitHub_Users_Emails.write_to_file-fc2d658a5c", "question": "Writes the user emails to file.", "scope": {"holder": "LLNL-scraper", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LionelAuroux/pyrser:pyrser/parsing/base.py:BasicParser.begin_tag#18525", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyrser/parsing/base.py", "symbol": "BasicParser.begin_tag"}, "source_id": "csn:LionelAuroux/pyrser:pyrser/parsing/base.py:BasicParser.begin_tag#18525"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LionelAuroux/pyrser/blob/f153a97ef2b6bf915a1ed468c0252a9a59b754d5/pyrser/parsing/base.py#L151-L155", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LionelAuroux-pyrser:BasicParser.begin_tag-e8bef76bd6", "question": "Save the current index under the given name.", "scope": {"holder": "LionelAuroux-pyrser", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LionelAuroux/pyrser:pyrser/type_system/scope.py:Scope.last#18473", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyrser/type_system/scope.py", "symbol": "Scope.last"}, "source_id": "csn:LionelAuroux/pyrser:pyrser/type_system/scope.py:Scope.last#18473"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LionelAuroux/pyrser/blob/f153a97ef2b6bf915a1ed468c0252a9a59b754d5/pyrser/type_system/scope.py#L338-L341", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LionelAuroux-pyrser:Scope.last-e63eb9ae74", "question": "Retrieve the last Signature ordered by mangling descendant", "scope": {"holder": "LionelAuroux-pyrser", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LordSputnik/mutagen:mutagen/apev2.py:APEValue#17388", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "mutagen/apev2.py", "symbol": "APEValue"}, "source_id": "csn:LordSputnik/mutagen:mutagen/apev2.py:APEValue#17388"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LordSputnik/mutagen/blob/38e62c8dc35c72b16554f5dbe7c0fde91acc3411/mutagen/apev2.py#L460-L482", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LordSputnik-mutagen:APEValue-c8b60189de", "question": "APEv2 tag value factory.\n\n Use this if you need to specify the value's type manually. Binary\n and text data are automatically detected by APEv2.__setitem__.", "scope": {"holder": "LordSputnik-mutagen", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LordSputnik/mutagen:mutagen/mp4.py:MP4Tags.__update_parents#17377", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "mutagen/mp4.py", "symbol": "MP4Tags.__update_parents"}, "source_id": "csn:LordSputnik/mutagen:mutagen/mp4.py:MP4Tags.__update_parents#17377"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LordSputnik/mutagen/blob/38e62c8dc35c72b16554f5dbe7c0fde91acc3411/mutagen/mp4.py#L447-L459", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LordSputnik-mutagen:MP4Tags.__update_parents-020231e413", "question": "Update all parent atoms with the new size.", "scope": {"holder": "LordSputnik-mutagen", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Loudr/asana-hub:asana_hub/action.py:Action.add_arguments#15127", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "asana_hub/action.py", "symbol": "Action.add_arguments"}, "source_id": "csn:Loudr/asana-hub:asana_hub/action.py:Action.add_arguments#15127"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Loudr/asana-hub/blob/af996ce890ed23d8ede5bf68dcd318e3438829cb/asana_hub/action.py#L27-L62", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Loudr-asana-hub:Action.add_arguments-ad6e162612", "question": "Add arguments to the parser for collection in app.args.\n\n Args:\n parser:\n `argparse.ArgumentParser`. Parser.\n Arguments added here are server on\n self.args.", "scope": {"holder": "Loudr-asana-hub", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Loudr/asana-hub:asana_hub/tool.py:ToolApp.get_saved_task_data#15140", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "asana_hub/tool.py", "symbol": "ToolApp.get_saved_task_data"}, "source_id": "csn:Loudr/asana-hub:asana_hub/tool.py:ToolApp.get_saved_task_data#15140"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Loudr/asana-hub/blob/af996ce890ed23d8ede5bf68dcd318e3438829cb/asana_hub/tool.py#L262-L282", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Loudr-asana-hub:ToolApp.get_saved_task_data-7ea18ce669", "question": "Returns task data from local data.\n\n Args:\n task:\n `int`. Asana task number.", "scope": {"holder": "Loudr-asana-hub", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LudovicRousseau/PyKCS11:PyKCS11/__init__.py:CK_OBJECT_HANDLE.to_dict#11133", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "PyKCS11/__init__.py", "symbol": "CK_OBJECT_HANDLE.to_dict"}, "source_id": "csn:LudovicRousseau/PyKCS11:PyKCS11/__init__.py:CK_OBJECT_HANDLE.to_dict#11133"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LudovicRousseau/PyKCS11/blob/76ccd8741af2ea193aaf1ca29dfedfa412c134fe/PyKCS11/__init__.py#L129-L155", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LudovicRousseau-PyKCS11:CK_OBJECT_HANDLE.to_dict-c534018c2a", "question": "convert the fields of the object into a dictionnary", "scope": {"holder": "LudovicRousseau-PyKCS11", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LudovicRousseau/PyKCS11:PyKCS11/__init__.py:Session.digestSession#11158", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "PyKCS11/__init__.py", "symbol": "Session.digestSession"}, "source_id": "csn:LudovicRousseau/PyKCS11:PyKCS11/__init__.py:Session.digestSession#11158"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LudovicRousseau/PyKCS11/blob/76ccd8741af2ea193aaf1ca29dfedfa412c134fe/PyKCS11/__init__.py#L975-L985", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LudovicRousseau-PyKCS11:Session.digestSession-cb30c32aba", "question": "C_DigestInit/C_DigestUpdate/C_DigestKey/C_DigestFinal\n\n :param mecha: the digesting mechanism to be used\n (use `MechanismSHA1` for `CKM_SHA_1`)\n :type mecha: :class:`Mechanism`\n :return: A :class:`DigestSession` object\n :rtype: DigestSession", "scope": {"holder": "LudovicRousseau-PyKCS11", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LuminosoInsight/luminoso-api-client-python:luminoso_api/v4_client.py:LuminosoClient._get_default_account#9049", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "luminoso_api/v4_client.py", "symbol": "LuminosoClient._get_default_account"}, "source_id": "csn:LuminosoInsight/luminoso-api-client-python:luminoso_api/v4_client.py:LuminosoClient._get_default_account#9049"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LuminosoInsight/luminoso-api-client-python/blob/3bedf2a454aee39214c11fbf556ead3eecc27881/luminoso_api/v4_client.py#L389-L403", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LuminosoInsight-luminoso-api-client-python:LuminosoClient._get_default_account-5fbbe5b72a", "question": "Get the ID of an account you can use to access projects.", "scope": {"holder": "LuminosoInsight-luminoso-api-client-python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:LuminosoInsight/luminoso-api-client-python:luminoso_api/v5_cli.py:_print_csv#9054", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "luminoso_api/v5_cli.py", "symbol": "_print_csv"}, "source_id": "csn:LuminosoInsight/luminoso-api-client-python:luminoso_api/v5_cli.py:_print_csv#9054"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/LuminosoInsight/luminoso-api-client-python/blob/3bedf2a454aee39214c11fbf556ead3eecc27881/luminoso_api/v5_cli.py#L43-L51", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:LuminosoInsight-luminoso-api-client-python:_print_csv-6efed6e092", "question": "Print a JSON list of JSON objects in CSV format.", "scope": {"holder": "LuminosoInsight-luminoso-api-client-python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ManageIQ/manageiq-api-client-python:src/manageiq_client/api.py:Action._get_entity_from_href#11874", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/manageiq_client/api.py", "symbol": "Action._get_entity_from_href"}, "source_id": "csn:ManageIQ/manageiq-api-client-python:src/manageiq_client/api.py:Action._get_entity_from_href#11874"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ManageIQ/manageiq-api-client-python/blob/e0c8884929e45766c2835bc7dcf4e78b0794248f/src/manageiq_client/api.py#L703-L724", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ManageIQ-manageiq-api-client-python:Action._get_entity_from_href-cffa72314a", "question": "Returns entity in correct collection.\n\n If the \"href\" value in result doesn't match the current collection,\n try to find the collection that the \"href\" refers to.", "scope": {"holder": "ManageIQ-manageiq-api-client-python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ManageIQ/manageiq-api-client-python:src/manageiq_client/api.py:Collection.raw_filter#11872", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/manageiq_client/api.py", "symbol": "Collection.raw_filter"}, "source_id": "csn:ManageIQ/manageiq-api-client-python:src/manageiq_client/api.py:Collection.raw_filter#11872"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ManageIQ/manageiq-api-client-python/blob/e0c8884929e45766c2835bc7dcf4e78b0794248f/src/manageiq_client/api.py#L339-L349", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ManageIQ-manageiq-api-client-python:Collection.raw_filter-429c7959c2", "question": "Sends all filters to the API.\n\n No fancy, just a wrapper. Any advanced functionality shall be implemented as another method.\n\n Args:\n filters: List of filters (strings)\n\n Returns: :py:class:`SearchResult`", "scope": {"holder": "ManageIQ-manageiq-api-client-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:MartinHjelmare/leicacam:leicacam/async_cam.py:AsyncCAM.close#13447", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "leicacam/async_cam.py", "symbol": "AsyncCAM.close"}, "source_id": "csn:MartinHjelmare/leicacam:leicacam/async_cam.py:AsyncCAM.close#13447"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MartinHjelmare/leicacam/blob/1df37bccd34884737d3b5e169fae71dd2f21f1e2/leicacam/async_cam.py#L97-L101", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MartinHjelmare-leicacam:AsyncCAM.close-572f03ab21", "question": "Close stream.", "scope": {"holder": "MartinHjelmare-leicacam", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:MartinHjelmare/leicacam:leicacam/cam.py:CAM.receive#13482", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "leicacam/cam.py", "symbol": "CAM.receive"}, "source_id": "csn:MartinHjelmare/leicacam:leicacam/cam.py:CAM.receive#13482"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MartinHjelmare/leicacam/blob/1df37bccd34884737d3b5e169fae71dd2f21f1e2/leicacam/cam.py#L222-L229", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MartinHjelmare-leicacam:CAM.receive-16672ded8f", "question": "Receive message from socket interface as list of OrderedDict.", "scope": {"holder": "MartinHjelmare-leicacam", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:MaxStrange/AudioSegment:audiosegment.py:AudioSegment._execute_sox_cmd#9385", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "audiosegment.py", "symbol": "AudioSegment._execute_sox_cmd"}, "source_id": "csn:MaxStrange/AudioSegment:audiosegment.py:AudioSegment._execute_sox_cmd#9385"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MaxStrange/AudioSegment/blob/1daefb8de626ddff3ff7016697c3ad31d262ecd6/audiosegment.py#L619-L663", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MaxStrange-AudioSegment:AudioSegment._execute_sox_cmd-63bb463677", "question": "Executes a Sox command in a platform-independent manner.\n\n `cmd` must be a format string that includes {inputfile} and {outputfile}.", "scope": {"holder": "MaxStrange-AudioSegment", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:MaxStrange/AudioSegment:algorithms/asa.py:_get_corresponding_offsets#9412", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "algorithms/asa.py", "symbol": "_get_corresponding_offsets"}, "source_id": "csn:MaxStrange/AudioSegment:algorithms/asa.py:_get_corresponding_offsets#9412"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MaxStrange/AudioSegment/blob/1daefb8de626ddff3ff7016697c3ad31d262ecd6/algorithms/asa.py#L605-L613", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MaxStrange-AudioSegment:_get_corresponding_offsets-83d0cf1886", "question": "Gets the offsets that occur as close as possible to the onsets in the given onset-front.", "scope": {"holder": "MaxStrange-AudioSegment", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MeirKriheli/django-bidi-utils:bidiutils/templatetags/bidiutils_tags.py:add_direction#19952", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "bidiutils/templatetags/bidiutils_tags.py", "symbol": "add_direction"}, "source_id": "csn:MeirKriheli/django-bidi-utils:bidiutils/templatetags/bidiutils_tags.py:add_direction#19952"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MeirKriheli/django-bidi-utils/blob/48a8c481fe728fbccf486582999e79454195036e/bidiutils/templatetags/bidiutils_tags.py#L10-L42", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MeirKriheli-django-bidi-utils:add_direction-f18ffc8120", "question": "Adds direction to the element\n\n :arguments:\n arg\n * rtl_only: Add the direction only in case of a\n right-to-left language (default)\n * both: add the direction in both case\n * ltr_only: Add the direction only in case of a\n left-to-right language\n\n {{image_name|add_direction}} when image_name is 'start_arrow.png'\n results in 'start_arrow_rtl.png' in case of RTL language, and\n 'start_arrow.png' or 'start_arrow_ltr.png' depends on `arg` value.", "scope": {"holder": "MeirKriheli-django-bidi-utils", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MeirKriheli/django-bidi-utils:bidiutils/context_processors.py:bidi#20065", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "bidiutils/context_processors.py", "symbol": "bidi"}, "source_id": "csn:MeirKriheli/django-bidi-utils:bidiutils/context_processors.py:bidi#20065"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MeirKriheli/django-bidi-utils/blob/48a8c481fe728fbccf486582999e79454195036e/bidiutils/context_processors.py#L3-L33", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MeirKriheli-django-bidi-utils:bidi-84cd369ea2", "question": "Adds to the context BiDi related variables\n\n LANGUAGE_DIRECTION -- Direction of current language ('ltr' or 'rtl')\n LANGUAGE_START -- Start of language layout ('right' for rtl, 'left'\n for 'ltr')\n LANGUAGE_END -- End of language layout ('left' for rtl, 'right'\n for 'ltr')\n LANGUAGE_MARKER -- Language marker entity ('‏' for rtl, '&lrm'\n for ltr)", "scope": {"holder": "MeirKriheli-django-bidi-utils", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MeirKriheli/python-bidi:bidi/algorithm.py:apply_mirroring#14321", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "bidi/algorithm.py", "symbol": "apply_mirroring"}, "source_id": "csn:MeirKriheli/python-bidi:bidi/algorithm.py:apply_mirroring#14321"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MeirKriheli/python-bidi/blob/a0e265bb465c1b7ad628487991e33b5ebe364641/bidi/algorithm.py#L580-L596", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MeirKriheli-python-bidi:apply_mirroring-67dd9d0b49", "question": "Applies L4: mirroring\n\n See: http://unicode.org/reports/tr9/#L4", "scope": {"holder": "MeirKriheli-python-bidi", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:MeirKriheli/python-bidi:bidi/algorithm.py:get_embedding_levels#14313", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "bidi/algorithm.py", "symbol": "get_embedding_levels"}, "source_id": "csn:MeirKriheli/python-bidi:bidi/algorithm.py:get_embedding_levels#14313"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MeirKriheli/python-bidi/blob/a0e265bb465c1b7ad628487991e33b5ebe364641/bidi/algorithm.py#L153-L181", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MeirKriheli-python-bidi:get_embedding_levels-7121b64673", "question": "Get the paragraph base embedding level and direction,\n set the storage to the array of chars", "scope": {"holder": "MeirKriheli-python-bidi", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Microsoft/botbuilder-python:libraries/botbuilder-ai/botbuilder/ai/luis/activity_util.py:ActivityUtil.create_trace#3617", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "libraries/botbuilder-ai/botbuilder/ai/luis/activity_util.py", "symbol": "ActivityUtil.create_trace"}, "source_id": "csn:Microsoft/botbuilder-python:libraries/botbuilder-ai/botbuilder/ai/luis/activity_util.py:ActivityUtil.create_trace#3617"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Microsoft/botbuilder-python/blob/274663dd91c811bae6ac4488915ba5880771b0a7/libraries/botbuilder-ai/botbuilder/ai/luis/activity_util.py#L16-L69", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Microsoft-botbuilder-python:ActivityUtil.create_trace-1a0674ca90", "question": "Creates a trace activity based on this activity.\n\n :param turn_activity:\n :type turn_activity: Activity\n :param name: The value to assign to the trace activity's property.\n :type name: str\n :param value: The value to assign to the trace activity's property., defaults to None\n :param value: object, optional\n :param value_type: The value to assign to the trace activity's property, defaults to None\n :param value_type: str, optional\n :param label: The value to assign to the trace activity's property, defaults to None\n :param label: str, optional\n :return: The created trace activity.\n :rtype: Activity", "scope": {"holder": "Microsoft-botbuilder-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Microsoft/botbuilder-python:libraries/botbuilder-azure/botbuilder/azure/cosmosdb_storage.py:CosmosDbStorage.__create_dict#3636", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "libraries/botbuilder-azure/botbuilder/azure/cosmosdb_storage.py", "symbol": "CosmosDbStorage.__create_dict"}, "source_id": "csn:Microsoft/botbuilder-python:libraries/botbuilder-azure/botbuilder/azure/cosmosdb_storage.py:CosmosDbStorage.__create_dict#3636"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Microsoft/botbuilder-python/blob/274663dd91c811bae6ac4488915ba5880771b0a7/libraries/botbuilder-azure/botbuilder/azure/cosmosdb_storage.py#L176-L189", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Microsoft-botbuilder-python:CosmosDbStorage.__create_dict-e4ddac00b6", "question": "Return the dict of a StoreItem.\n\n This eliminates non_magic attributes and the e_tag.\n\n :param si:\n :return dict:", "scope": {"holder": "Microsoft-botbuilder-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:MinchinWeb/minchin.text:minchin/text.py:centered#19652", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "minchin/text.py", "symbol": "centered"}, "source_id": "csn:MinchinWeb/minchin.text:minchin/text.py:centered#19652"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MinchinWeb/minchin.text/blob/4d136039561892c3adab49fe81b2f246e5a1507b/minchin/text.py#L85-L95", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MinchinWeb-minchin.text:centered-f29aa3dbf6", "question": "Takes a string, centres it, and pads it on both sides", "scope": {"holder": "MinchinWeb-minchin.text", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MinchinWeb/minchin.text:minchin/text.py:query_yes_no#19654", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "minchin/text.py", "symbol": "query_yes_no"}, "source_id": "csn:MinchinWeb/minchin.text:minchin/text.py:query_yes_no#19654"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MinchinWeb/minchin.text/blob/4d136039561892c3adab49fe81b2f246e5a1507b/minchin/text.py#L106-L139", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MinchinWeb-minchin.text:query_yes_no-c399a21d13", "question": "Ask a yes/no question via raw_input() and return their answer.\n\n \"question\" is a string that is presented to the user.\n \"default\" is the presumed answer if the user just hits .\n It must be \"yes\" (the default), \"no\" or None (meaning\n an answer is required of the user).\n\n The return value is one of Answers.YES or Answers.NO.\n\n Copied (and modified) from\n http://stackoverflow.com/questions/3041986/python-command-line-yes-no-input", "scope": {"holder": "MinchinWeb-minchin.text", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MisterY/price-database:pricedb/config.py:Config.__create_user_config#9893", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pricedb/config.py", "symbol": "Config.__create_user_config"}, "source_id": "csn:MisterY/price-database:pricedb/config.py:Config.__create_user_config#9893"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MisterY/price-database/blob/b4fd366b7763891c690fe3000b8840e656da023e/pricedb/config.py#L74-L89", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MisterY-price-database:Config.__create_user_config-be8f8832f9", "question": "Copy the config template into user's directory", "scope": {"holder": "MisterY-price-database", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MisterY/price-database:pricedb/app.py:PriceDbApplication.get_price_repository#9908", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pricedb/app.py", "symbol": "PriceDbApplication.get_price_repository"}, "source_id": "csn:MisterY/price-database:pricedb/app.py:PriceDbApplication.get_price_repository#9908"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MisterY/price-database/blob/b4fd366b7763891c690fe3000b8840e656da023e/pricedb/app.py#L210-L216", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MisterY-price-database:PriceDbApplication.get_price_repository-d8eefbfb16", "question": "Price repository", "scope": {"holder": "MisterY-price-database", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MitchellChu/torndsession:demos/memory_session.py:DeleteHandler.get#17586", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "demos/memory_session.py", "symbol": "DeleteHandler.get"}, "source_id": "csn:MitchellChu/torndsession:demos/memory_session.py:DeleteHandler.get#17586"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MitchellChu/torndsession/blob/dd08554c06f47d33396a0a4485f53d0522961155/demos/memory_session.py#L53-L65", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MitchellChu-torndsession:DeleteHandler.get-49726e2201", "question": "Please don't do this in production environments.", "scope": {"holder": "MitchellChu-torndsession", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MomsFriendlyRobotCompany/fake_rpi:fake_rpi/Adafruit.py:LSM303.read#8292", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "fake_rpi/Adafruit.py", "symbol": "LSM303.read"}, "source_id": "csn:MomsFriendlyRobotCompany/fake_rpi:fake_rpi/Adafruit.py:LSM303.read#8292"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MomsFriendlyRobotCompany/fake_rpi/blob/eb2f640f19f74db9082bfbf8aed524e3ccba6d4f/fake_rpi/Adafruit.py#L24-L34", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MomsFriendlyRobotCompany-fake_rpi:LSM303.read-77bc37e008", "question": "Since there isn't a real sensor connected, read() creates random\n\t\tdata.", "scope": {"holder": "MomsFriendlyRobotCompany-fake_rpi", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MuhammedHasan/sklearn_utils:sklearn_utils/preprocessing/dict_input.py:DictInput.transform#16527", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sklearn_utils/preprocessing/dict_input.py", "symbol": "DictInput.transform"}, "source_id": "csn:MuhammedHasan/sklearn_utils:sklearn_utils/preprocessing/dict_input.py:DictInput.transform#16527"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MuhammedHasan/sklearn_utils/blob/337c3b7a27f4921d12da496f66a2b83ef582b413/sklearn_utils/preprocessing/dict_input.py#L29-L39", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MuhammedHasan-sklearn_utils:DictInput.transform-819d61c867", "question": ":param X: features.", "scope": {"holder": "MuhammedHasan-sklearn_utils", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MuhammedHasan/sklearn_utils:sklearn_utils/utils/skutils_io.py:SkUtilsIO.to_json#16532", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sklearn_utils/utils/skutils_io.py", "symbol": "SkUtilsIO.to_json"}, "source_id": "csn:MuhammedHasan/sklearn_utils:sklearn_utils/utils/skutils_io.py:SkUtilsIO.to_json#16532"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MuhammedHasan/sklearn_utils/blob/337c3b7a27f4921d12da496f66a2b83ef582b413/sklearn_utils/utils/skutils_io.py#L48-L57", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MuhammedHasan-sklearn_utils:SkUtilsIO.to_json-3e75640edf", "question": "Reads dataset to csv.\n\n :param X: dataset as list of dict.\n :param y: labels.", "scope": {"holder": "MuhammedHasan-sklearn_utils", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Music-Moo/music2storage:music2storage/connection.py:ConnectionHandler.use_music_service#16528", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "music2storage/connection.py", "symbol": "ConnectionHandler.use_music_service"}, "source_id": "csn:Music-Moo/music2storage:music2storage/connection.py:ConnectionHandler.use_music_service#16528"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Music-Moo/music2storage/blob/de12b9046dd227fc8c1512b5060e7f5fcd8b0ee2/music2storage/connection.py#L16-L34", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Music-Moo-music2storage:ConnectionHandler.use_music_service-d293629472", "question": "Sets the current music service to service_name.\n\n :param str service_name: Name of the music service\n :param str api_key: Optional API key if necessary", "scope": {"holder": "Music-Moo-music2storage", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Music-Moo/music2storage:music2storage/__init__.py:Music2Storage.add_to_queue#16509", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "music2storage/__init__.py", "symbol": "Music2Storage.add_to_queue"}, "source_id": "csn:Music-Moo/music2storage:music2storage/__init__.py:Music2Storage.add_to_queue#16509"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Music-Moo/music2storage/blob/de12b9046dd227fc8c1512b5060e7f5fcd8b0ee2/music2storage/__init__.py#L36-L48", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Music-Moo-music2storage:Music2Storage.add_to_queue-7c668dba00", "question": "Adds an URL to the download queue.\n\n :param str url: URL to the music service track", "scope": {"holder": "Music-Moo-music2storage", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:MycroftAI/padaos:padaos.py:IntentContainer._create_regex#12161", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "padaos.py", "symbol": "IntentContainer._create_regex"}, "source_id": "csn:MycroftAI/padaos:padaos.py:IntentContainer._create_regex#12161"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/MycroftAI/padaos/blob/c7fb3d72fefbe552963c5a5a0f606e1f2fde2968/padaos.py#L104-L112", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:MycroftAI-padaos:IntentContainer._create_regex-98fedd09a6", "question": "Create regex and return. If error occurs returns None.", "scope": {"holder": "MycroftAI-padaos", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:NLeSC/pycoeman:pycoeman/utils_execution.py:apply_argument_parser#15247", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pycoeman/utils_execution.py", "symbol": "apply_argument_parser"}, "source_id": "csn:NLeSC/pycoeman:pycoeman/utils_execution.py:apply_argument_parser#15247"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/NLeSC/pycoeman/blob/246d517b55cb98eb46f69aae453492cf9dd9d5af/pycoeman/utils_execution.py#L51-L57", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:NLeSC-pycoeman:apply_argument_parser-4f8aedcf63", "question": "Apply the argument parser.", "scope": {"holder": "NLeSC-pycoeman", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:NarrativeScience/lsi:src/lsi/utils/hosts.py:HostEntry._get_attrib#19339", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/lsi/utils/hosts.py", "symbol": "HostEntry._get_attrib"}, "source_id": "csn:NarrativeScience/lsi:src/lsi/utils/hosts.py:HostEntry._get_attrib#19339"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/NarrativeScience/lsi/blob/7d901b03fdb1a34ef795e5412bfe9685d948e32d/src/lsi/utils/hosts.py#L146-L179", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:NarrativeScience-lsi:HostEntry._get_attrib-b255fda2b2", "question": "Given an attribute name, looks it up on the entry. Names that\n start with ``tags.`` are looked up in the ``tags`` dictionary.\n\n :param attr: Name of attribute to look up.\n :type attr: ``str``\n :param convert_to_str: Convert result to a string.\n :type convert_to_str: ``bool``\n\n :rtype: ``object``", "scope": {"holder": "NarrativeScience-lsi", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:NarrativeScience/lsi:src/lsi/lsi.py:_run_ssh#19387", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "src/lsi/lsi.py", "symbol": "_run_ssh"}, "source_id": "csn:NarrativeScience/lsi:src/lsi/lsi.py:_run_ssh#19387"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/NarrativeScience/lsi/blob/7d901b03fdb1a34ef795e5412bfe9685d948e32d/src/lsi/lsi.py#L195-L350", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:NarrativeScience-lsi:_run_ssh-c35c6a3a50", "question": "Lets the user choose which instance to SSH into.\n\n :param entries: The list of host entries.\n :type entries: [:py:class:`HostEntry`]\n :param username: The SSH username to use. Defaults to current user.\n :type username: ``str`` or ``NoneType``\n :param idfile: The identity file to use. Optional.\n :type idfile: ``str`` or ``NoneType``\n :param no_prompt: Whether to disable confirmation for SSH command.\n :type no_prompt: ``bool``\n :param command: SSH command to run on matching instances.\n :type command: ``str`` or ``NoneType``\n :param show: Instance attributes to show in addition to defaults.\n :type show: ``NoneType`` or ``list`` of ``str``\n :param only: If not ``None``, will *only* show these attributes.\n :type only: ``NoneType`` or ``list`` of ``str``\n :param sort_by: What to sort columns by. By default, sort by 'name'.\n :type sort_by: ``str``\n :param limit: At most how many results to show.\n :type limit: ``int`` or ``NoneType``\n :param num: If not None, choose the given entry from within filters.\n :type num: ``int`` or ``NoneType``\n :param random: If true, choose a random entry from within filters.\n :type random: ``bool``", "scope": {"holder": "NarrativeScience-lsi", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:NateFerrero/oauth2lib:oauth2lib/provider.py:AuthorizationProvider.get_authorization_code#14084", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "oauth2lib/provider.py", "symbol": "AuthorizationProvider.get_authorization_code"}, "source_id": "csn:NateFerrero/oauth2lib:oauth2lib/provider.py:AuthorizationProvider.get_authorization_code#14084"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/NateFerrero/oauth2lib/blob/d161b010f8a596826050a09e5e94d59443cc12d9/oauth2lib/provider.py#L205-L268", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:NateFerrero-oauth2lib:AuthorizationProvider.get_authorization_code-9c982f7df5", "question": "Generate authorization code HTTP response.\n\n :param response_type: Desired response type. Must be exactly \"code\".\n :type response_type: str\n :param client_id: Client ID.\n :type client_id: str\n :param redirect_uri: Client redirect URI.\n :type redirect_uri: str\n :rtype: requests.Response", "scope": {"holder": "NateFerrero-oauth2lib", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:NateFerrero/oauth2lib:oauth2lib/provider.py:Provider._make_json_response#14083", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "oauth2lib/provider.py", "symbol": "Provider._make_json_response"}, "source_id": "csn:NateFerrero/oauth2lib:oauth2lib/provider.py:Provider._make_json_response#14083"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/NateFerrero/oauth2lib/blob/d161b010f8a596826050a09e5e94d59443cc12d9/oauth2lib/provider.py#L61-L80", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:NateFerrero-oauth2lib:Provider._make_json_response-cb7434bd3a", "question": "Return a response object from the given JSON data.\n\n :param data: Data to JSON-encode.\n :type data: mixed\n :param headers: Dict of headers to include in the requests.\n :type headers: dict\n :param status_code: HTTP status code.\n :type status_code: int\n :rtype: requests.Response", "scope": {"holder": "NateFerrero-oauth2lib", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Nekmo/amazon-dash:amazon_dash/config.py:Config.read#6531", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "amazon_dash/config.py", "symbol": "Config.read"}, "source_id": "csn:Nekmo/amazon-dash:amazon_dash/config.py:Config.read#6531"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Nekmo/amazon-dash/blob/0e2bdc24ff8ea32cecb2f5f54f5cc1c0f99c197b/amazon_dash/config.py#L227-L240", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Nekmo-amazon-dash:Config.read-2c5d8fb71a", "question": "Parse and validate the config file. The read data is accessible as a dictionary in this instance\n\n :return: None", "scope": {"holder": "Nekmo-amazon-dash", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Nekmo/amazon-dash:amazon_dash/listener.py:Listener.run#6550", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "amazon_dash/listener.py", "symbol": "Listener.run"}, "source_id": "csn:Nekmo/amazon-dash:amazon_dash/listener.py:Listener.run#6550"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Nekmo/amazon-dash/blob/0e2bdc24ff8ea32cecb2f5f54f5cc1c0f99c197b/amazon_dash/listener.py#L152-L159", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Nekmo-amazon-dash:Listener.run-89057f767e", "question": "Start daemon mode\n\n :param bool root_allowed: Only used for ExecuteCmd\n :return: loop", "scope": {"holder": "Nekmo-amazon-dash", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Nekroze/greencard:greencard/greencard.py:card#19716", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "greencard/greencard.py", "symbol": "card"}, "source_id": "csn:Nekroze/greencard:greencard/greencard.py:card#19716"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Nekroze/greencard/blob/30fe7eba5742c31b666027e31f33aaa641699857/greencard/greencard.py#L9-L19", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Nekroze-greencard:card-21eb5e421e", "question": "A decorator for providing a unittesting function/method with every card in\n a librarian card library database when it is called.", "scope": {"holder": "Nekroze-greencard", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Nekroze/greencard:greencard/greencard.py:library#19717", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "greencard/greencard.py", "symbol": "library"}, "source_id": "csn:Nekroze/greencard:greencard/greencard.py:library#19717"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Nekroze/greencard/blob/30fe7eba5742c31b666027e31f33aaa641699857/greencard/greencard.py#L22-L32", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Nekroze-greencard:library-067ffd3dcc", "question": "A decorator for providing a unittest with a library and have it called only\n once.", "scope": {"holder": "Nekroze-greencard", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Nic30/hwt:hwt/hdl/assignment.py:Assignment._cut_off_drivers_of#9677", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "hwt/hdl/assignment.py", "symbol": "Assignment._cut_off_drivers_of"}, "source_id": "csn:Nic30/hwt:hwt/hdl/assignment.py:Assignment._cut_off_drivers_of#9677"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Nic30/hwt/blob/8cbb399e326da3b22c233b98188a9d08dec057e6/hwt/hdl/assignment.py#L69-L77", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Nic30-hwt:Assignment._cut_off_drivers_of-6e151838a5", "question": "Cut off statements which are driver of specified signal", "scope": {"holder": "Nic30-hwt", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Nic30/hwt:hwt/hdl/switchContainer.py:SwitchContainer._condHasEffect#9696", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "hwt/hdl/switchContainer.py", "symbol": "SwitchContainer._condHasEffect"}, "source_id": "csn:Nic30/hwt:hwt/hdl/switchContainer.py:SwitchContainer._condHasEffect#9696"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Nic30/hwt/blob/8cbb399e326da3b22c233b98188a9d08dec057e6/hwt/hdl/switchContainer.py#L228-L251", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Nic30-hwt:SwitchContainer._condHasEffect-ad4445d38f", "question": ":return: True if statements in branches has different effect", "scope": {"holder": "Nic30-hwt", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Numergy/yoda:yoda/adapter/abstract.py:Abstract.execute#17898", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "yoda/adapter/abstract.py", "symbol": "Abstract.execute"}, "source_id": "csn:Numergy/yoda:yoda/adapter/abstract.py:Abstract.execute#17898"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Numergy/yoda/blob/109f0e9441130488b0155f05883ef6531cf46ee9/yoda/adapter/abstract.py#L37-L62", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Numergy-yoda:Abstract.execute-cb0f4d61ce", "question": "Execute command with os.popen and return output.", "scope": {"holder": "Numergy-yoda", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Numergy/yoda:yoda/subcommand/status.py:Status.print_workspace#17903", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "yoda/subcommand/status.py", "symbol": "Status.print_workspace"}, "source_id": "csn:Numergy/yoda:yoda/subcommand/status.py:Status.print_workspace#17903"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Numergy/yoda/blob/109f0e9441130488b0155f05883ef6531cf46ee9/yoda/subcommand/status.py#L56-L65", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Numergy-yoda:Status.print_workspace-2b1ddfe63d", "question": "Print workspace status.", "scope": {"holder": "Numergy-yoda", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ONSdigital/sdc-cryptography:sdc/crypto/key_store.py:KeyStore.get_key_for_purpose_and_type#13433", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sdc/crypto/key_store.py", "symbol": "KeyStore.get_key_for_purpose_and_type"}, "source_id": "csn:ONSdigital/sdc-cryptography:sdc/crypto/key_store.py:KeyStore.get_key_for_purpose_and_type#13433"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ONSdigital/sdc-cryptography/blob/846feb2b27b1c62d35ff2c290c05abcead68b23c/sdc/crypto/key_store.py#L62-L72", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ONSdigital-sdc-cryptography:KeyStore.get_key_for_purpose_and_type-25f922d56c", "question": "Gets a list of keys that match the purpose and key_type, and returns the first key in that list\n Note, if there are many keys that match the criteria, the one you get back will be random from that list\n :returns: A key object that matches the criteria", "scope": {"holder": "ONSdigital-sdc-cryptography", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ONSdigital/sdc-cryptography:sdc/crypto/decrypter.py:decrypt#13431", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sdc/crypto/decrypter.py", "symbol": "decrypt"}, "source_id": "csn:ONSdigital/sdc-cryptography:sdc/crypto/decrypter.py:decrypt#13431"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ONSdigital/sdc-cryptography/blob/846feb2b27b1c62d35ff2c290c05abcead68b23c/sdc/crypto/decrypter.py#L6-L25", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ONSdigital-sdc-cryptography:decrypt-dbab51df03", "question": "This decrypts the provided jwe token, then decodes resulting jwt token and returns\n the payload.\n\n :param str token: The jwe token.\n :param key_store: The key store.\n :param str key_purpose: Context for the key.\n :param int leeway: Extra allowed time in seconds after expiration to account for clock skew.\n :return: The decrypted payload.", "scope": {"holder": "ONSdigital-sdc-cryptography", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:OldhamMade/exemelopy:exemelopy/__init__.py:XMLEncoder.to_string#22078", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "exemelopy/__init__.py", "symbol": "XMLEncoder.to_string"}, "source_id": "csn:OldhamMade/exemelopy:exemelopy/__init__.py:XMLEncoder.to_string#22078"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/OldhamMade/exemelopy/blob/5f5141b169e61a5b6912146a995917f5d862ee9c/exemelopy/__init__.py#L52-L66", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:OldhamMade-exemelopy:XMLEncoder.to_string-051193b4dc", "question": "Encodes the stored ``data`` to XML and returns a\n ``string``.\n\n Setting ``indent`` to ``False`` will forego any pretty-printing\n and return a condensed value.\n\n Setting ``declaration`` to ``False`` will skip inserting the\n XML declaration.", "scope": {"holder": "OldhamMade-exemelopy", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:OldhamMade/exemelopy:exemelopy/__init__.py:XMLEncoder.to_xml#22079", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "exemelopy/__init__.py", "symbol": "XMLEncoder.to_xml"}, "source_id": "csn:OldhamMade/exemelopy:exemelopy/__init__.py:XMLEncoder.to_xml#22079"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/OldhamMade/exemelopy/blob/5f5141b169e61a5b6912146a995917f5d862ee9c/exemelopy/__init__.py#L68-L75", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:OldhamMade-exemelopy:XMLEncoder.to_xml-2fae15444e", "question": "Encodes the stored ``data`` to XML and returns\n an ``lxml.etree`` value.", "scope": {"holder": "OldhamMade-exemelopy", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:OpenKMIP/PyKMIP:kmip/core/messages/payloads/activate.py:ActivateRequestPayload.validate#7196", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "kmip/core/messages/payloads/activate.py", "symbol": "ActivateRequestPayload.validate"}, "source_id": "csn:OpenKMIP/PyKMIP:kmip/core/messages/payloads/activate.py:ActivateRequestPayload.validate#7196"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/OpenKMIP/PyKMIP/blob/b51c5b044bd05f8c85a1d65d13a583a4d8fc1b0e/kmip/core/messages/payloads/activate.py#L95-L103", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:OpenKMIP-PyKMIP:ActivateRequestPayload.validate-8bf883f4f3", "question": "Error check the attributes of the ActivateRequestPayload object.", "scope": {"holder": "OpenKMIP-PyKMIP", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:OpenKMIP/PyKMIP:kmip/services/server/engine.py:KmipEngine._get_attribute_from_managed_object#7050", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "kmip/services/server/engine.py", "symbol": "KmipEngine._get_attribute_from_managed_object"}, "source_id": "csn:OpenKMIP/PyKMIP:kmip/services/server/engine.py:KmipEngine._get_attribute_from_managed_object#7050"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/OpenKMIP/PyKMIP/blob/b51c5b044bd05f8c85a1d65d13a583a4d8fc1b0e/kmip/services/server/engine.py#L627-L719", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:OpenKMIP-PyKMIP:KmipEngine._get_attribute_from_managed_object-8f4bcf1342", "question": "Get the attribute value from the kmip.pie managed object.", "scope": {"holder": "OpenKMIP-PyKMIP", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:OpenMath/py-openmath:openmath/convert.py:Converter._deprecated_register#17819", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "openmath/convert.py", "symbol": "Converter._deprecated_register"}, "source_id": "csn:OpenMath/py-openmath:openmath/convert.py:Converter._deprecated_register#17819"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/OpenMath/py-openmath/blob/4906aa9ccf606f533675c28823772e07c30fd220/openmath/convert.py#L242-L250", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:OpenMath-py-openmath:Converter._deprecated_register-848345e331", "question": "This is a shorthand for:\n\n ``self.register_to_python(om_cd, om_name, to_py)``\n ``self.register_to_openmath(py_class, to_om)``", "scope": {"holder": "OpenMath-py-openmath", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:OpenMath/py-openmath:openmath/convert_pickle.py:cls_build#17869", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "openmath/convert_pickle.py", "symbol": "cls_build"}, "source_id": "csn:OpenMath/py-openmath:openmath/convert_pickle.py:cls_build#17869"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/OpenMath/py-openmath/blob/4906aa9ccf606f533675c28823772e07c30fd220/openmath/convert_pickle.py#L331-L389", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:OpenMath-py-openmath:cls_build-d530b95eb0", "question": "Apply the setstate protocol to initialize `inst` from `state`.\n\n INPUT:\n\n - ``inst`` -- a raw instance of a class\n - ``state`` -- the state to restore; typically a dictionary mapping attribute names to their values\n\n EXAMPLES::\n\n >>> from openmath.convert_pickle import cls_build\n >>> class A(object): pass\n >>> inst = A.__new__(A)\n >>> state = {\"foo\": 1, \"bar\": 4}\n >>> inst2 = cls_build(inst,state)\n >>> inst is inst2\n True\n >>> inst.foo\n 1\n >>> inst.bar\n 4", "scope": {"holder": "OpenMath-py-openmath", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Patreon/multiget-cache-py:multiget_cache/base_cache_wrapper.py:BaseCacheWrapper.delete#13436", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "multiget_cache/base_cache_wrapper.py", "symbol": "BaseCacheWrapper.delete"}, "source_id": "csn:Patreon/multiget-cache-py:multiget_cache/base_cache_wrapper.py:BaseCacheWrapper.delete#13436"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Patreon/multiget-cache-py/blob/824ec4809c97cc7e0035810bd9fefd1262de3318/multiget_cache/base_cache_wrapper.py#L55-L60", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Patreon-multiget-cache-py:BaseCacheWrapper.delete-2516784610", "question": "Remove the key from the request cache and from memcache.", "scope": {"holder": "Patreon-multiget-cache-py", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Patreon/multiget-cache-py:multiget_cache/function_tools.py:map_arguments_to_objects#13435", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "multiget_cache/function_tools.py", "symbol": "map_arguments_to_objects"}, "source_id": "csn:Patreon/multiget-cache-py:multiget_cache/function_tools.py:map_arguments_to_objects#13435"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Patreon/multiget-cache-py/blob/824ec4809c97cc7e0035810bd9fefd1262de3318/multiget_cache/function_tools.py#L105-L124", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Patreon-multiget-cache-py:map_arguments_to_objects-dd30ce50b8", "question": ":param kwargs: kwargs used to call the multiget function\n :param objects: objects returned from the inner function\n :param object_key: field or set of fields that map to the kwargs provided\n :param object_tuple_key: A temporary shortcut until we allow dot.path traversal for object_key.\n Will call getattr(getattr(result, join_table_name), object_key)\n :param argument_key: field or set of fields that map to the objects provided\n :param result_value: Limit the fields returned to this field or set of fields (none = whole object)\n :param default_result: If the inner function returned none for a set of parameters, default to this\n :return:", "scope": {"holder": "Patreon-multiget-cache-py", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PierreRaybaut/PyQtdoc:setup.py:get_data_files#17775", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "setup.py", "symbol": "get_data_files"}, "source_id": "csn:PierreRaybaut/PyQtdoc:setup.py:get_data_files#17775"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PierreRaybaut/PyQtdoc/blob/71dce1cf11d958f72db2f5b9519d3540569b7720/setup.py#L8-L14", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PierreRaybaut-PyQtdoc:get_data_files-b5079c3e68", "question": "Return data files in directory *dirname*", "scope": {"holder": "PierreRaybaut-PyQtdoc", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Pirionfr/pyLinky:pylinky/client.py:LinkyClient._get_data#9068", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pylinky/client.py", "symbol": "LinkyClient._get_data"}, "source_id": "csn:Pirionfr/pyLinky:pylinky/client.py:LinkyClient._get_data#9068"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Pirionfr/pyLinky/blob/4372496bfcdd95ccfd2f017634cf02b38a2d6fd1/pylinky/client.py#L76-L126", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Pirionfr-pyLinky:LinkyClient._get_data-f60d4fe78f", "question": "Get data.", "scope": {"holder": "Pirionfr-pyLinky", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Pirionfr/pyLinky:pylinky/client.py:LinkyClient._post_login_page#9067", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pylinky/client.py", "symbol": "LinkyClient._post_login_page"}, "source_id": "csn:Pirionfr/pyLinky:pylinky/client.py:LinkyClient._post_login_page#9067"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Pirionfr/pyLinky/blob/4372496bfcdd95ccfd2f017634cf02b38a2d6fd1/pylinky/client.py#L55-L74", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Pirionfr-pyLinky:LinkyClient._post_login_page-85cefbd8db", "question": "Login to enedis.", "scope": {"holder": "Pirionfr-pyLinky", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PolicyStat/docx2html:docx2html/core.py:_strip_tag#14162", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "docx2html/core.py", "symbol": "_strip_tag"}, "source_id": "csn:PolicyStat/docx2html:docx2html/core.py:_strip_tag#14162"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PolicyStat/docx2html/blob/2dc4afd1e3a3f2f0b357d0bff903eb58bcc94429/docx2html/core.py#L1344-L1350", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PolicyStat-docx2html:_strip_tag-d0c4b975f6", "question": "Remove all tags that have the tag name ``tag``", "scope": {"holder": "PolicyStat-docx2html", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PolicyStat/docx2html:docx2html/core.py:get_style_dict#14153", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "docx2html/core.py", "symbol": "get_style_dict"}, "source_id": "csn:PolicyStat/docx2html:docx2html/core.py:get_style_dict#14153"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PolicyStat/docx2html/blob/2dc4afd1e3a3f2f0b357d0bff903eb58bcc94429/docx2html/core.py#L674-L729", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PolicyStat-docx2html:get_style_dict-c22c46cfc7", "question": "Some things that are considered lists are actually supposed to be H tags\n (h1, h2, etc.) These can be denoted by their styleId", "scope": {"holder": "PolicyStat-docx2html", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Polyconseil/devpi-metawheel:devpi_metawheel/main.py:devpiserver_on_upload#13735", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "devpi_metawheel/main.py", "symbol": "devpiserver_on_upload"}, "source_id": "csn:Polyconseil/devpi-metawheel:devpi_metawheel/main.py:devpiserver_on_upload#13735"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Polyconseil/devpi-metawheel/blob/8794f7b74c406979aa4a35fa1cd28e1e9c8864ae/devpi_metawheel/main.py#L17-L52", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Polyconseil-devpi-metawheel:devpiserver_on_upload-c26515fbce", "question": "called when a file is uploaded to a private stage for\n a projectname/version. link.entry.file_exists() may be false because\n a more recent revision deleted the file (and files are not revisioned).\n NOTE that this hook is currently NOT called for the implicit \"caching\"\n uploads to the pypi mirror.\n\n If the uploaded file is a wheel and is the latest version on this index,\n store its metadata in json file at the root of index/+f/ directory.\n With the standard config with nginx, nginx will directly serve this file.", "scope": {"holder": "Polyconseil-devpi-metawheel", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:PyCQA/pylint:pylint/utils/ast_walker.py:ASTWalker.add_checker#5879", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pylint/utils/ast_walker.py", "symbol": "ASTWalker.add_checker"}, "source_id": "csn:PyCQA/pylint:pylint/utils/ast_walker.py:ASTWalker.add_checker#5879"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PyCQA/pylint/blob/2bf5c61a3ff6ae90613b81679de42c0f19aea600/pylint/utils/ast_walker.py#L27-L55", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PyCQA-pylint:ASTWalker.add_checker-5c44da7a9a", "question": "walk to the checker's dir and collect visit and leave methods", "scope": {"holder": "PyCQA-pylint", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PyCQA/pylint:pylint/checkers/raw_metrics.py:RawMetricsChecker.process_tokens#5775", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pylint/checkers/raw_metrics.py", "symbol": "RawMetricsChecker.process_tokens"}, "source_id": "csn:PyCQA/pylint:pylint/checkers/raw_metrics.py:RawMetricsChecker.process_tokens#5775"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PyCQA/pylint/blob/2bf5c61a3ff6ae90613b81679de42c0f19aea600/pylint/checkers/raw_metrics.py#L77-L84", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PyCQA-pylint:RawMetricsChecker.process_tokens-7f7c421111", "question": "update stats", "scope": {"holder": "PyCQA-pylint", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PyFilesystem/s3fs:fs_s3fs/_s3fs.py:S3File.factory#10222", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "fs_s3fs/_s3fs.py", "symbol": "S3File.factory"}, "source_id": "csn:PyFilesystem/s3fs:fs_s3fs/_s3fs.py:S3File.factory#10222"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PyFilesystem/s3fs/blob/1c5e3a1b6abbb9dff91ea7fc4cec7353798cd536/fs_s3fs/_s3fs.py#L65-L69", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PyFilesystem-s3fs:S3File.factory-4fa100ab85", "question": "Create a S3File backed with a temporary file.", "scope": {"holder": "PyFilesystem-s3fs", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:PyFilesystem/s3fs:fs_s3fs/_s3fs.py:_make_repr#10220", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "fs_s3fs/_s3fs.py", "symbol": "_make_repr"}, "source_id": "csn:PyFilesystem/s3fs:fs_s3fs/_s3fs.py:_make_repr#10220"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PyFilesystem/s3fs/blob/1c5e3a1b6abbb9dff91ea7fc4cec7353798cd536/fs_s3fs/_s3fs.py#L34-L58", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PyFilesystem-s3fs:_make_repr-306be9867b", "question": "Generate a repr string.\n\n Positional arguments should be the positional arguments used to\n construct the class. Keyword arguments should consist of tuples of\n the attribute value and default. If the value is the default, then\n it won't be rendered in the output.\n\n Here's an example::\n\n def __repr__(self):\n return make_repr('MyClass', 'foo', name=(self.name, None))\n\n The output of this would be something line ``MyClass('foo',\n name='Will')``.", "scope": {"holder": "PyFilesystem-s3fs", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PyProphet/pyprophet:pyprophet/main.py:backpropagate#12237", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyprophet/main.py", "symbol": "backpropagate"}, "source_id": "csn:PyProphet/pyprophet:pyprophet/main.py:backpropagate#12237"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PyProphet/pyprophet/blob/f546ad171750cd7685afbde6785fe71f82cadb35/pyprophet/main.py#L247-L257", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PyProphet-pyprophet:backpropagate-f486c81fd2", "question": "Backpropagate multi-run peptide and protein scores to single files", "scope": {"holder": "PyProphet-pyprophet", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:PyProphet/pyprophet:pyprophet/main.py:peptide#12232", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyprophet/main.py", "symbol": "peptide"}, "source_id": "csn:PyProphet/pyprophet:pyprophet/main.py:peptide#12232"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PyProphet/pyprophet/blob/f546ad171750cd7685afbde6785fe71f82cadb35/pyprophet/main.py#L144-L154", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PyProphet-pyprophet:peptide-e9af82a648", "question": "Infer peptides and conduct error-rate estimation in different contexts.", "scope": {"holder": "PyProphet-pyprophet", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:PythonSanSebastian/docstamp:docstamp/data_source.py:GoogleData.getCSV#19164", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "docstamp/data_source.py", "symbol": "GoogleData.getCSV"}, "source_id": "csn:PythonSanSebastian/docstamp:docstamp/data_source.py:GoogleData.getCSV#19164"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PythonSanSebastian/docstamp/blob/b43808f2e15351b0b2f0b7eade9c7ef319c9e646/docstamp/data_source.py#L11-L39", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PythonSanSebastian-docstamp:GoogleData.getCSV-4e8e2fc1f1", "question": "Returns\n -------\n filename: str", "scope": {"holder": "PythonSanSebastian-docstamp", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:PythonSanSebastian/docstamp:docstamp/file_utils.py:cleanup#19106", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "docstamp/file_utils.py", "symbol": "cleanup"}, "source_id": "csn:PythonSanSebastian/docstamp:docstamp/file_utils.py:cleanup#19106"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/PythonSanSebastian/docstamp/blob/b43808f2e15351b0b2f0b7eade9c7ef319c9e646/docstamp/file_utils.py#L118-L129", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:PythonSanSebastian-docstamp:cleanup-7d14da8065", "question": "Remove the files in workdir that have the given extension.\n\n Parameters\n ----------\n workdir:\n Folder path from where to clean the files.\n\n extension: str\n File extension without the dot, e.g., 'txt'", "scope": {"holder": "PythonSanSebastian-docstamp", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Qiskit/qiskit-terra:qiskit/converters/ast_to_dag.py:AstInterpreter._create_dag_op#3743", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "qiskit/converters/ast_to_dag.py", "symbol": "AstInterpreter._create_dag_op"}, "source_id": "csn:Qiskit/qiskit-terra:qiskit/converters/ast_to_dag.py:AstInterpreter._create_dag_op#3743"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Qiskit/qiskit-terra/blob/d4f58d903bc96341b816f7c35df936d6421267d1/qiskit/converters/ast_to_dag.py#L306-L377", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Qiskit-qiskit-terra:AstInterpreter._create_dag_op-b15961168d", "question": "Create a DAG node out of a parsed AST op node.\n\n Args:\n name (str): operation name to apply to the dag.\n params (list): op parameters\n qargs (list(QuantumRegister, int)): qubits to attach to\n\n Raises:\n QiskitError: if encountering a non-basis opaque gate", "scope": {"holder": "Qiskit-qiskit-terra", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Qiskit/qiskit-terra:qiskit/circuit/quantumcircuit.py:QuantumCircuit._bind_parameter#3914", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "qiskit/circuit/quantumcircuit.py", "symbol": "QuantumCircuit._bind_parameter"}, "source_id": "csn:Qiskit/qiskit-terra:qiskit/circuit/quantumcircuit.py:QuantumCircuit._bind_parameter#3914"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Qiskit/qiskit-terra/blob/d4f58d903bc96341b816f7c35df936d6421267d1/qiskit/circuit/quantumcircuit.py#L725-L728", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Qiskit-qiskit-terra:QuantumCircuit._bind_parameter-bb54a16251", "question": "Assigns a parameter value to matching instructions in-place.", "scope": {"holder": "Qiskit-qiskit-terra", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:RRZE-HPC/kerncraft:kerncraft/models/benchmark.py:Benchmark.analyze#12077", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "kerncraft/models/benchmark.py", "symbol": "Benchmark.analyze"}, "source_id": "csn:RRZE-HPC/kerncraft:kerncraft/models/benchmark.py:Benchmark.analyze#12077"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/RRZE-HPC/kerncraft/blob/c60baf8043e4da8d8d66da7575021c2f4c6c78af/kerncraft/models/benchmark.py#L343-L486", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:RRZE-HPC-kerncraft:Benchmark.analyze-57963ff82c", "question": "Run analysis.", "scope": {"holder": "RRZE-HPC-kerncraft", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:RRZE-HPC/kerncraft:kerncraft/kernel.py:KernelCode.iaca_analysis#12054", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "kerncraft/kernel.py", "symbol": "KernelCode.iaca_analysis"}, "source_id": "csn:RRZE-HPC/kerncraft:kerncraft/kernel.py:KernelCode.iaca_analysis#12054"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/RRZE-HPC/kerncraft/blob/c60baf8043e4da8d8d66da7575021c2f4c6c78af/kerncraft/kernel.py#L1514-L1535", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:RRZE-HPC-kerncraft:KernelCode.iaca_analysis-362649ef89", "question": "Run an IACA analysis and return its outcome.\n\n *asm_block* controls how the to-be-marked block is chosen. \"auto\" (default) results in\n the largest block, \"manual\" results in interactive and a number in the according block.\n\n *pointer_increment* is the number of bytes the pointer is incremented after the loop or\n - 'auto': automatic detection, RuntimeError is raised in case of failure\n - 'auto_with_manual_fallback': automatic detection, fallback to manual input\n - 'manual': prompt user", "scope": {"holder": "RRZE-HPC-kerncraft", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:RazerM/yourls-python:yourls/core.py:YOURLSAPIMixin.db_stats#13505", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "yourls/core.py", "symbol": "YOURLSAPIMixin.db_stats"}, "source_id": "csn:RazerM/yourls-python:yourls/core.py:YOURLSAPIMixin.db_stats#13505"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/RazerM/yourls-python/blob/716845562a2bbb430de3c379c9481b195e451ccf/yourls/core.py#L166-L181", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:RazerM-yourls-python:YOURLSAPIMixin.db_stats-fc65a012dd", "question": "Get database statistics.\n\n Returns:\n DBStats: Total clicks and links statistics.\n\n Raises:\n requests.exceptions.HTTPError: Generic HTTP Error", "scope": {"holder": "RazerM-yourls-python", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:RazerM/yourls-python:yourls/core.py:YOURLSAPIMixin.stats#13504", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "yourls/core.py", "symbol": "YOURLSAPIMixin.stats"}, "source_id": "csn:RazerM/yourls-python:yourls/core.py:YOURLSAPIMixin.stats#13504"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/RazerM/yourls-python/blob/716845562a2bbb430de3c379c9481b195e451ccf/yourls/core.py#L121-L164", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:RazerM-yourls-python:YOURLSAPIMixin.stats-8cca43918e", "question": "Get stats about links.\n\n Parameters:\n filter: 'top', 'bottom', 'rand', or 'last'.\n limit: Number of links to return from filter.\n start: Optional start number.\n\n Returns:\n Tuple containing list of ShortenedURLs and DBStats.\n\n Example:\n\n .. code-block:: python\n\n links, stats = yourls.stats(filter='top', limit=10)\n\n Raises:\n ValueError: Incorrect value for filter parameter.\n requests.exceptions.HTTPError: Generic HTTP Error", "scope": {"holder": "RazerM-yourls-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:SeabornGames/Logger:seaborn_logger/logger.py:SeabornLogger.findCaller#19813", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "seaborn_logger/logger.py", "symbol": "SeabornLogger.findCaller"}, "source_id": "csn:SeabornGames/Logger:seaborn_logger/logger.py:SeabornLogger.findCaller#19813"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/SeabornGames/Logger/blob/fb8b1700557aaea8d3216bd4c4df33c302bece7f/seaborn_logger/logger.py#L45-L72", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:SeabornGames-Logger:SeabornLogger.findCaller-f01ee29b63", "question": "Find the stack frame of the caller so that we can note the source\n file name, line number and function name.", "scope": {"holder": "SeabornGames-Logger", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:SeabornGames/Logger:seaborn_logger/logger.py:setup_file_logging#19809", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "seaborn_logger/logger.py", "symbol": "setup_file_logging"}, "source_id": "csn:SeabornGames/Logger:seaborn_logger/logger.py:setup_file_logging#19809"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/SeabornGames/Logger/blob/fb8b1700557aaea8d3216bd4c4df33c302bece7f/seaborn_logger/logger.py#L289-L323", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:SeabornGames-Logger:setup_file_logging-283045562e", "question": "This will setup logging for a single file but can be called more than once\n LOG LEVELS are \"CRITICAL\", \"ERROR\", \"INFO\", \"DEBUG\"\n :param log_filename: str of the file location\n :param log_file_level str of the log level to use on this file\n :param str_format: str of the logging format\n :param date_format: str of the date format\n :param log_restart: bool if True the log file will be deleted first\n :param log_history: bool if True will save another log file in a folder\n called history with the datetime\n :param formatter: logging.Format instance to use\n :param log_filter: logging.filter instance to add to handler\n :param silence_modules list of str of modules to silence\n :return: None", "scope": {"holder": "SeabornGames-Logger", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:SectorLabs/django-postgres-extra:psqlextra/datastructures.py:ConditionalJoin.add_condition#4851", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "psqlextra/datastructures.py", "symbol": "ConditionalJoin.add_condition"}, "source_id": "csn:SectorLabs/django-postgres-extra:psqlextra/datastructures.py:ConditionalJoin.add_condition#4851"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/SectorLabs/django-postgres-extra/blob/eef2ed5504d225858d4e4f5d77a838082ca6053e/psqlextra/datastructures.py#L17-L28", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:SectorLabs-django-postgres-extra:ConditionalJoin.add_condition-e83603d980", "question": "Adds an extra condition to this join.\n\n Arguments:\n field:\n The field that the condition will apply to.\n\n value:\n The value to compare.", "scope": {"holder": "SectorLabs-django-postgres-extra", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:SectorLabs/django-postgres-extra:psqlextra/compiler.py:PostgresInsertCompiler._get_model_field#4838", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "psqlextra/compiler.py", "symbol": "PostgresInsertCompiler._get_model_field"}, "source_id": "csn:SectorLabs/django-postgres-extra:psqlextra/compiler.py:PostgresInsertCompiler._get_model_field#4838"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/SectorLabs/django-postgres-extra/blob/eef2ed5504d225858d4e4f5d77a838082ca6053e/psqlextra/compiler.py#L240-L266", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:SectorLabs-django-postgres-extra:PostgresInsertCompiler._get_model_field-27ebe5dc0d", "question": "Gets the field on a model with the specified name.\n\n Arguments:\n name:\n The name of the field to look for.\n\n This can be both the actual field name, or\n the name of the column, both will work :)\n\n Returns:\n The field with the specified name or None if\n no such field exists.", "scope": {"holder": "SectorLabs-django-postgres-extra", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:SiLab-Bonn/pixel_clusterizer:pixel_clusterizer/clusterizer.py:HitClusterizer.add_cluster_field#18546", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pixel_clusterizer/clusterizer.py", "symbol": "HitClusterizer.add_cluster_field"}, "source_id": "csn:SiLab-Bonn/pixel_clusterizer:pixel_clusterizer/clusterizer.py:HitClusterizer.add_cluster_field#18546"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/SiLab-Bonn/pixel_clusterizer/blob/d2c8c3072fb03ebb7c6a3e8c57350fbbe38efd4d/pixel_clusterizer/clusterizer.py#L241-L252", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:SiLab-Bonn-pixel_clusterizer:HitClusterizer.add_cluster_field-bfb64b9159", "question": "Adds a field or a list of fields to the cluster result array. Has to be defined as a numpy dtype entry, e.g.: ('parameter', ' [xmin, xmax, ymin, ymax]\n\n >>> Index.deinterleave([0, 10, 1, 11])\n [0, 1, 10, 11]\n\n >>> Index.deinterleave([0, 1, 2, 10, 11, 12])\n [0, 10, 1, 11, 2, 12]", "scope": {"holder": "Toblerity-rtree", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ToucanToco/toucan-data-sdk:toucan_data_sdk/sdk.py:ToucanDataSdk.read_entry#9159", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "toucan_data_sdk/sdk.py", "symbol": "ToucanDataSdk.read_entry"}, "source_id": "csn:ToucanToco/toucan-data-sdk:toucan_data_sdk/sdk.py:ToucanDataSdk.read_entry#9159"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ToucanToco/toucan-data-sdk/blob/c3ca874e1b64f4bdcc2edda750a72d45d1561d8a/toucan_data_sdk/sdk.py#L127-L137", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ToucanToco-toucan-data-sdk:ToucanDataSdk.read_entry-87965c4c4b", "question": "Args:\n file_name (str):\n\n Returns:\n pd.DataFrame:", "scope": {"holder": "ToucanToco-toucan-data-sdk", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ToucanToco/toucan-data-sdk:toucan_data_sdk/sdk.py:extract#9157", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "toucan_data_sdk/sdk.py", "symbol": "extract"}, "source_id": "csn:ToucanToco/toucan-data-sdk:toucan_data_sdk/sdk.py:extract#9157"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ToucanToco/toucan-data-sdk/blob/c3ca874e1b64f4bdcc2edda750a72d45d1561d8a/toucan_data_sdk/sdk.py#L194-L213", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ToucanToco-toucan-data-sdk:extract-623730ea3b", "question": "Args:\n data (str | byte):\n\n Returns:\n dict: Dict[str, DataFrame]", "scope": {"holder": "ToucanToco-toucan-data-sdk", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Trim21/sdu_bkjws:sdu_bkjws/__init__.py:SduBkjws._aodata#19560", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sdu_bkjws/__init__.py", "symbol": "SduBkjws._aodata"}, "source_id": "csn:Trim21/sdu_bkjws:sdu_bkjws/__init__.py:SduBkjws._aodata#19560"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Trim21/sdu_bkjws/blob/87e74c84e6a20b2abd20905e2381f0a572398f6c/sdu_bkjws/__init__.py#L54-L86", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Trim21-sdu_bkjws:SduBkjws._aodata-42e8029479", "question": "生成用于post的数据\n\n :param echo: a int to check is response is write\n :type echo: int\n :param columns: 所有columns列名组成的list\n :type columns: list\n :param xnxq: str\n :type xnxq: string\n :param final_exam: 是否期末考试\n :rtype: bool\n :return: a valid data for post to get data", "scope": {"holder": "Trim21-sdu_bkjws", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Trim21/sdu_bkjws:sdu_bkjws/__init__.py:SduBkjws.get_lesson#19562", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "sdu_bkjws/__init__.py", "symbol": "SduBkjws.get_lesson"}, "source_id": "csn:Trim21/sdu_bkjws:sdu_bkjws/__init__.py:SduBkjws.get_lesson#19562"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Trim21/sdu_bkjws/blob/87e74c84e6a20b2abd20905e2381f0a572398f6c/sdu_bkjws/__init__.py#L135-L163", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Trim21-sdu_bkjws:SduBkjws.get_lesson-99889b4e44", "question": "获取课表,返回一个列表,包含所有课表对象\n\n :return: list of lessons\n :rtype: list[dict]", "scope": {"holder": "Trim21-sdu_bkjws", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Tsjerk/Insane:insane/lipids.py:Lipid.build#17296", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "insane/lipids.py", "symbol": "Lipid.build"}, "source_id": "csn:Tsjerk/Insane:insane/lipids.py:Lipid.build#17296"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Tsjerk/Insane/blob/b73f08910ddb0b66597b20ff75ecee7f65f4ecf6/insane/lipids.py#L90-L144", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Tsjerk-Insane:Lipid.build-cf6d07e309", "question": "Build/return a list of [(bead, x, y, z), ...]", "scope": {"holder": "Tsjerk-Insane", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Tsjerk/Insane:insane/utils.py:iter_resource#17294", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "insane/utils.py", "symbol": "iter_resource"}, "source_id": "csn:Tsjerk/Insane:insane/utils.py:iter_resource#17294"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Tsjerk/Insane/blob/b73f08910ddb0b66597b20ff75ecee7f65f4ecf6/insane/utils.py#L25-L34", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Tsjerk-Insane:iter_resource-79fa442df3", "question": "Return a stream for a given resource file in the module.\n\n The resource file has to be part of the module and its filenane given\n relative to the module.", "scope": {"holder": "Tsjerk-Insane", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:UCBerkeleySETI/blimpy:blimpy/file_wrapper.py:FilReader._find_blob_start#11779", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "blimpy/file_wrapper.py", "symbol": "FilReader._find_blob_start"}, "source_id": "csn:UCBerkeleySETI/blimpy:blimpy/file_wrapper.py:FilReader._find_blob_start#11779"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/UCBerkeleySETI/blimpy/blob/b8822d3e3e911944370d84371a91fa0c29e9772e/blimpy/file_wrapper.py#L624-L639", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:UCBerkeleySETI-blimpy:FilReader._find_blob_start-66c06f9bec", "question": "Find first blob from selection.", "scope": {"holder": "UCBerkeleySETI-blimpy", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:UCBerkeleySETI/blimpy:blimpy/waterfall.py:Waterfall.__write_to_hdf5_heavy#11791", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "blimpy/waterfall.py", "symbol": "Waterfall.__write_to_hdf5_heavy"}, "source_id": "csn:UCBerkeleySETI/blimpy:blimpy/waterfall.py:Waterfall.__write_to_hdf5_heavy#11791"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/UCBerkeleySETI/blimpy/blob/b8822d3e3e911944370d84371a91fa0c29e9772e/blimpy/waterfall.py#L309-L404", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:UCBerkeleySETI-blimpy:Waterfall.__write_to_hdf5_heavy-dcb644cef3", "question": "Write data to HDF5 file.\n\n Args:\n filename_out (str): Name of output file", "scope": {"holder": "UCBerkeleySETI-blimpy", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:UDST/pandana:pandana/network.py:Network.aggregate#10805", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pandana/network.py", "symbol": "Network.aggregate"}, "source_id": "csn:UDST/pandana:pandana/network.py:Network.aggregate#10805"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/UDST/pandana/blob/961a7ef8d3b0144b190cb60bbd61845fca6fb314/pandana/network.py#L274-L336", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:UDST-pandana:Network.aggregate-e087a871e1", "question": "Aggregate information for every source node in the network - this is\n really the main purpose of this library. This allows you to touch\n the data specified by calling set and perform some aggregation on it\n within the specified distance. For instance, summing the population\n within 1000 meters.\n\n Parameters\n ----------\n distance : float\n The maximum distance to aggregate data within. 'distance' can\n represent any impedance unit that you have set as your edge\n weight. This will usually be a distance unit in meters however\n if you have customized the impedance this could be in other\n units such as utility or time etc.\n type : string\n The type of aggregation, can be one of \"ave\", \"sum\", \"std\",\n \"count\", and now \"min\", \"25pct\", \"median\", \"75pct\", and \"max\" will\n compute the associated quantiles. (Quantiles are computed by\n sorting so might be slower than the others.)\n decay : string\n The type of decay to apply, which makes things that are further\n away count less in the aggregation - must be one of \"linear\",\n \"exponential\" or \"flat\" (which means no decay). Linear is the\n fastest computation to perform. When performing an \"ave\",\n the decay is typically \"flat\"\n imp_name : string, optional\n The impedance name to use for the aggregation on this network.\n Must be one of the impedance names passed in the constructor of\n this object. If not specified, there must be only one impedance\n passed in the constructor, which will be used.\n name : string, optional\n The variable to aggregate. This variable will have been created\n and named by a call to set. If not specified, the default\n variable name will be used so that the most recent call to set\n without giving a name will be the variable used.\n\n Returns\n -------\n agg : Pandas Series\n Returns a Pandas Series for every origin node in the network,\n with the index which is the same as the node_ids passed to the\n init method and the values are the aggregations for each source\n node in the network.", "scope": {"holder": "UDST-pandana", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:UDST/pandana:pandana/loaders/osm.py:build_node_query#10814", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pandana/loaders/osm.py", "symbol": "build_node_query"}, "source_id": "csn:UDST/pandana:pandana/loaders/osm.py:build_node_query#10814"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/UDST/pandana/blob/961a7ef8d3b0144b190cb60bbd61845fca6fb314/pandana/loaders/osm.py#L120-L157", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:UDST-pandana:build_node_query-53fe220c9b", "question": "Build the string for a node-based OSM query.\n\n Parameters\n ----------\n lat_min, lng_min, lat_max, lng_max : float\n tags : str or list of str, optional\n Node tags that will be used to filter the search.\n See http://wiki.openstreetmap.org/wiki/Overpass_API/Language_Guide\n for information about OSM Overpass queries\n and http://wiki.openstreetmap.org/wiki/Map_Features\n for a list of tags.\n\n Returns\n -------\n query : str", "scope": {"holder": "UDST-pandana", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:URXtech/cmph-cffi:cmph/__init__.py:MPH.lookup#19280", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "cmph/__init__.py", "symbol": "MPH.lookup"}, "source_id": "csn:URXtech/cmph-cffi:cmph/__init__.py:MPH.lookup#19280"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/URXtech/cmph-cffi/blob/85298572e51675cd0c7ef1052ed9989b0e57f0cc/cmph/__init__.py#L252-L273", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:URXtech-cmph-cffi:MPH.lookup-8397716751", "question": "Generate hash code for a key from the Minimal Perfect Hash (MPH)\n\n Parameters\n ----------\n Key : object\n The item to generate a key for, this works best for keys that\n are strings, or can be transformed fairly directly into bytes\n\n Returns : int\n The code for the given item", "scope": {"holder": "URXtech-cmph-cffi", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:URXtech/cmph-cffi:cmph/__init__.py:generate_hash#19277", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "cmph/__init__.py", "symbol": "generate_hash"}, "source_id": "csn:URXtech/cmph-cffi:cmph/__init__.py:generate_hash#19277"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/URXtech/cmph-cffi/blob/85298572e51675cd0c7ef1052ed9989b0e57f0cc/cmph/__init__.py#L388-L567", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:URXtech-cmph-cffi:generate_hash-956208fecb", "question": "Generates a new Minimal Perfect Hash (MPH)\n\n Parameters\n ----------\n data : list, array-like, file-like\n The input that is used to generate the minimal perfect hash.\n\n Be aware, in most cases the input is expected to be distinct, and\n many of the algorithms benefit from the input being sorted.\n\n algorithm : string, optional\n {chd_ph (default), chd, bmz, bmz8, chm, brz, fch, bdz, bdz_ph}\n The algorithm to use in generating MPH's, choice of:\n\n chd / chd_ph - Compress Hash and Displace (default)\n (http://cmph.sourceforge.net/chd.html)\n - It is the fastest algorithm to build PHFs and MPHFs in linear\n time.\n - It generates the most compact PHFs and MPHFs we know of.\n - It can generate PHFs with a load factor up to 99 %.\n - It can be used to generate t-perfect hash functions. A\n t-perfect hash function allows at most t collisions in a given\n bin. It is a well-known fact that modern memories are\n organized as blocks which constitute transfer unit. Example of\n such blocks are cache lines for internal memory or sectors for\n hard disks. Thus, it can be very useful for devices that\n carry out I/O operations in blocks.\n - It is a two level scheme. It uses a first level hash function\n to split the key set in buckets of average size determined by\n a parameter b in the range [1,32]. In the second level it uses\n displacement values to resolve the collisions that have given\n rise to the buckets.\n - It can generate MPHFs that can be stored in approximately 2.07\n bits per key.\n - For a load factor equal to the maximum one that is achieved by\n the BDZ algorithm (81 %), the resulting PHFs are stored in\n approximately 1.40 bits per key.\n\n bdz - BDZ / BPZ algorithm\n (http://cmph.sourceforge.net/bdz.html)\n - It is very simple and efficient. It outperforms all others\n except CHD.\n - It constructs both PHFs and MPHFs in linear time.\n - The maximum load factor one can achieve for a PHF is 1/1.23.\n - It is based on acyclic random 3-graphs. A 3-graph is a\n generalization of a graph where each edge connects 3 vertices\n instead of only 2.\n - The resulting MPHFs are not order preserving.\n - The resulting MPHFs can be stored in only (2 + x)cn bits,\n where c should be larger than or equal to 1.23 and x is a\n constant larger than 0 (actually, x = 1/b and b is a parameter\n that should be larger than 2). For c = 1.23 and b = 8, the\n resulting functions are stored in approximately 2.6 bits per\n key.\n - For its maximum load factor (81 %), the resulting PHFs are\n stored in approximately 1.95 bits per key.\n\n bmz - Botelho, Menoti and Ziviani algorithm:\n (http://cmph.sourceforge.net/bdz.html)\n - Constructs MPHFs in linear time.\n - It is based on cyclic random graphs. This makes it faster than\n the CHM algorithm.\n - The resulting MPHFs are not order preserving.\n - The resulting MPHFs are more compact than the ones generated by\n the CHM algorithm and can be stored in 4cn bytes, where c is in\n the range [0.93,1.15].\n\n brz - BRZ algorithm:\n (http://cmph.sourceforge.net/brz.html)\n - A very fast external memory based algorithm for constructing\n minimal perfect hash functions for sets in the order of\n billions of keys.\n - It works in linear time.\n - The resulting MPHFs are not order preserving.\n - The resulting MPHFs can be stored using less than 8.0 bits per\n key.\n\n chm - Czech, Havas and Majewski algorithm:\n (http://cmph.sourceforge.net/chm.html)\n - Construct minimal MPHFs in linear time.\n - It is based on acyclic random graphs\n - The resulting MPHFs are order preserving.\n - The resulting MPHFs are stored in 4cn bytes, where c is greater\n than 2.\n\n fch - Fox, Chen and Heath algorithm:\n (http://cmph.sourceforge.net/chm.html)\n - Construct minimal perfect hash functions that require less than\n 4 bits per key to be stored.\n - The resulting MPHFs are very compact and very efficient at\n evaluation time\n - The algorithm is only efficient for small sets.\n - It is used as internal algorithm in the BRZ algorithm to\n efficiently solve larger problems and even so to generate MPHFs\n that require approximately 4.1 bits per key to be stored. For\n that, you just need to set the parameters -a to brz and -c to a\n value larger than or equal to 2.6.\n\n hash_fns : list {jenkins (default), count} optional\n Internal hash functions to use inside MPH generation functions,\n can be multiple fns as a list.\n\n chd_keys_per_bin : int [1 to 128], optional\n Set the number of keys per bin for a t-perfect hashing function. A\n t-perfect hash function allows at most t collisions in a given bin.\n This parameter applies only to the `chd` and `chd_ph` algorithms.\n Its value should be an integer in the range [1, 128].\n\n Default is 1\n\n chd_load_factor : float, optional\n The load factor used in the `chd_ph` algorithm\n\n fch_bits_per_key : int, optional\n The number of bits per key required in the FCH algorithm\n\n num_graph_vertices : int, optional\n The number of vertices in the graph for the algorithms BMZ and CHM\n\n brz_memory_size : int (default 8), optional\n Main memory availability (in MB) used in BRZ algorithm\n\n Default is 8Mb\n\n brz_temp_dir : string, optional\n Temporary directory used in BRZ algorithm\n\n brz_max_keys_per_bucket : int [64 to 175] (default 128), optional\n Used to make the maximal number of keys in a bucket lower than 256.\n In this case its value should be an integer in the range [64,175].\n Default is 128.\n\n bdz_precomputed_rank : int [3 to 10] (default 7), optional\n For BDZ it is used to determine the size of some precomputed rank\n information and its value should be an integer in the range [3,10].\n\n Default is 7.\n\n The larger is this value, the more compact are the resulting\n functions and the slower are them at evaluation time.\n\n chd_avg_keys_per_bucket : int [1 to 32] (default 4), optional\n For CHD and CHD_PH it is used to set the average number of keys per\n bucket and its value should be an integer in the range [1,32].\n\n Default is 4.\n\n The larger is this value, the slower is the construction of the\n functions.\n\n Returns\n -------\n MPH\n A wrapper object that represents a minimal perfect hash in memory\n\n Raises\n ------\n ValueError\n If arguments presented are incomplete, or incompatable\n RuntimeError\n If the MPH generation fails", "scope": {"holder": "URXtech-cmph-cffi", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Valuehorizon/valuehorizon-forex:forex/models.py:Currency.compute_return#21405", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "forex/models.py", "symbol": "Currency.compute_return"}, "source_id": "csn:Valuehorizon/valuehorizon-forex:forex/models.py:Currency.compute_return#21405"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Valuehorizon/valuehorizon-forex/blob/e921379ae6c9d07ddad87a1fd3b5bb8fdfc74cb8/forex/models.py#L59-L75", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Valuehorizon-valuehorizon-forex:Currency.compute_return-10ea9791d5", "question": "Compute the return of the currency between two dates", "scope": {"holder": "Valuehorizon-valuehorizon-forex", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Valuehorizon/valuehorizon-forex:forex/models.py:CurrencyPriceManager.generate_dataframe#21406", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "forex/models.py", "symbol": "CurrencyPriceManager.generate_dataframe"}, "source_id": "csn:Valuehorizon/valuehorizon-forex:forex/models.py:CurrencyPriceManager.generate_dataframe#21406"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Valuehorizon/valuehorizon-forex/blob/e921379ae6c9d07ddad87a1fd3b5bb8fdfc74cb8/forex/models.py#L81-L125", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Valuehorizon-valuehorizon-forex:CurrencyPriceManager.generate_dataframe-ed13cc5132", "question": "Generate a dataframe consisting of the currency prices (specified by symbols)\n from the start to end date", "scope": {"holder": "Valuehorizon-valuehorizon-forex", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:VingtCinq/python-resize-image:resizeimage/resizeimage.py:_height_is_big_enough#12992", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "resizeimage/resizeimage.py", "symbol": "_height_is_big_enough"}, "source_id": "csn:VingtCinq/python-resize-image:resizeimage/resizeimage.py:_height_is_big_enough#12992"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/VingtCinq/python-resize-image/blob/a4e645792ef30c5fcc558df6da6de18b1ecb95ea/resizeimage/resizeimage.py#L50-L53", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:VingtCinq-python-resize-image:_height_is_big_enough-e65dfc7fbb", "question": "Check that the image height is superior to `height`", "scope": {"holder": "VingtCinq-python-resize-image", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:VingtCinq/python-resize-image:resizeimage/resizeimage.py:resize_height#12997", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "resizeimage/resizeimage.py", "symbol": "resize_height"}, "source_id": "csn:VingtCinq/python-resize-image:resizeimage/resizeimage.py:resize_height#12997"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/VingtCinq/python-resize-image/blob/a4e645792ef30c5fcc558df6da6de18b1ecb95ea/resizeimage/resizeimage.py#L142-L162", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:VingtCinq-python-resize-image:resize_height-e139f734ce", "question": "Resize image according to size.\n image: a Pillow image instance\n size: an integer or a list or tuple of two integers [width, height]", "scope": {"holder": "VingtCinq-python-resize-image", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:VorskiImagineering/C3PO:c3po/mod/communicator.py:Communicator._clear_temp#20987", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "c3po/mod/communicator.py", "symbol": "Communicator._clear_temp"}, "source_id": "csn:VorskiImagineering/C3PO:c3po/mod/communicator.py:Communicator._clear_temp#20987"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/VorskiImagineering/C3PO/blob/e3e35835e5ac24158848afed4f905ca44ac3ae00/c3po/mod/communicator.py#L115-L125", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:VorskiImagineering-C3PO:Communicator._clear_temp-12802afa55", "question": "Clear temp directory from created csv and ods files during\n communicator operations.", "scope": {"holder": "VorskiImagineering-C3PO", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:VorskiImagineering/C3PO:c3po/converters/po_csv.py:_prepare_polib_files#20780", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "c3po/converters/po_csv.py", "symbol": "_prepare_polib_files"}, "source_id": "csn:VorskiImagineering/C3PO:c3po/converters/po_csv.py:_prepare_polib_files#20780"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/VorskiImagineering/C3PO/blob/e3e35835e5ac24158848afed4f905ca44ac3ae00/c3po/converters/po_csv.py#L61-L80", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:VorskiImagineering-C3PO:_prepare_polib_files-01e5a98124", "question": "Prepare polib file object for writing/reading from them.\n Create directories and write header if needed. For each language,\n ensure there's a translation file named \"filename\" in the correct place.\n Assumes (and creates) a directory structure:\n ///.", "scope": {"holder": "VorskiImagineering-C3PO", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:WGBH/wagtail-streamfieldtools:streamfield_tools/registry.py:RegisteredBlockStreamFieldRegistry.register_block#16545", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "streamfield_tools/registry.py", "symbol": "RegisteredBlockStreamFieldRegistry.register_block"}, "source_id": "csn:WGBH/wagtail-streamfieldtools:streamfield_tools/registry.py:RegisteredBlockStreamFieldRegistry.register_block#16545"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/WGBH/wagtail-streamfieldtools/blob/192f86845532742b0b7d432bef3987357833b8ed/streamfield_tools/registry.py#L51-L57", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:WGBH-wagtail-streamfieldtools:RegisteredBlockStreamFieldRegistry.register_block-73567e0473", "question": "Registers `block` to `block_type` in the registry.", "scope": {"holder": "WGBH-wagtail-streamfieldtools", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:WGBH/wagtail-streamfieldtools:streamfield_tools/blocks/struct_block.py:RenditionAwareStructBlock.get_context#16506", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "streamfield_tools/blocks/struct_block.py", "symbol": "RenditionAwareStructBlock.get_context"}, "source_id": "csn:WGBH/wagtail-streamfieldtools:streamfield_tools/blocks/struct_block.py:RenditionAwareStructBlock.get_context#16506"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/WGBH/wagtail-streamfieldtools/blob/192f86845532742b0b7d432bef3987357833b8ed/streamfield_tools/blocks/struct_block.py#L118-L123", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:WGBH-wagtail-streamfieldtools:RenditionAwareStructBlock.get_context-d9a4b359de", "question": "Ensure `image_rendition` is added to the global context.", "scope": {"holder": "WGBH-wagtail-streamfieldtools", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:WhoopInc/pytsdb:pytsdb.py:Connection.fetch_metric#16622", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pytsdb.py", "symbol": "Connection.fetch_metric"}, "source_id": "csn:WhoopInc/pytsdb:pytsdb.py:Connection.fetch_metric#16622"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/WhoopInc/pytsdb/blob/f053c2effa7a9175c5521e690dbb23700c6c0cfe/pytsdb.py#L20-L101", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:WhoopInc-pytsdb:Connection.fetch_metric-ea5e52bfb2", "question": "Fetch time series data from OpenTSDB\n\n Parameters:\n metric:\n A string representing a valid OpenTSDB metric.\n\n tags:\n A dict mapping tag names to tag values. Tag names and values are\n always strings.\n\n { 'user_id': '44' }\n\n start:\n A datetime.datetime-like object representing the start of the\n range to query over.\n\n end:\n A datetime.datetime-like object representing the end of the\n range to query over.\n\n aggregator:\n The function for merging multiple time series together. For\n example, if the \"user_id\" tag is not specified, this aggregator\n function is used to combine all heart rate time series into one\n time series. (Yes, this isn't very useful.)\n\n For queries that return only one time series, this parameter is\n not relevant.\n\n Valid values: \"sum\", \"min\", \"max\", \"avg\", \"dev\"\n\n See: http://opentsdb.net/docs/build/html/user_guide/query/aggregators.html\n\n downsampling:\n A relative time interval to \"downsample\". This isn't true\n downsampling; rather, if you specify a downsampling of \"5m\"\n (five minutes), OpenTSDB will split data into five minute\n intervals, and return one data point in the middle of each\n interval whose value is the average of all data points within\n that interval.\n\n Valid relative time values are strings of the following format:\n\n \"\"\n\n Valid time units: \"ms\", \"s\", \"m\", \"h\", \"d\", \"w\", \"n\", \"y\"\n\n Date and time format: http://opentsdb.net/docs/build/html/user_guide/query/dates.html\n\n ms_resolution:\n Whether or not to output data point timestamps in milliseconds\n or seconds. If this flag is false and there are multiple\n data points within a second, those data points will be down\n sampled using the query's aggregation function.\n\n Returns:\n A dict mapping timestamps to data points", "scope": {"holder": "WhoopInc-pytsdb", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:WhoopInc/pytsdb:pytsdb.py:Connection.fetch_sorted_metric#16623", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pytsdb.py", "symbol": "Connection.fetch_sorted_metric"}, "source_id": "csn:WhoopInc/pytsdb:pytsdb.py:Connection.fetch_sorted_metric#16623"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/WhoopInc/pytsdb/blob/f053c2effa7a9175c5521e690dbb23700c6c0cfe/pytsdb.py#L103-L110", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:WhoopInc-pytsdb:Connection.fetch_sorted_metric-44a26876fa", "question": "Fetch and sort time series data from OpenTSDB\n\n Takes the same parameters as `fetch_metric`, but returns a list of\n (timestamp, value) tuples sorted by timestamp.", "scope": {"holder": "WhoopInc-pytsdb", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:WoLpH/mt940:mt940/processors.py:date_fixup_pre_processor#7411", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "mt940/processors.py", "symbol": "date_fixup_pre_processor"}, "source_id": "csn:WoLpH/mt940:mt940/processors.py:date_fixup_pre_processor#7411"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/WoLpH/mt940/blob/fd71c88f7ec8372f4194f831f4c29c6f9bd9d0eb/mt940/processors.py#L17-L31", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:WoLpH-mt940:date_fixup_pre_processor-9b603a2d69", "question": "Replace illegal February 29, 30 dates with the last day of February.\n\n German banks use a variant of the 30/360 interest rate calculation,\n where each month has always 30 days even February. Python's datetime\n module won't accept such dates.", "scope": {"holder": "WoLpH-mt940", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:WoLpH/mt940:mt940/processors.py:mBank_set_tnr#7414", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "mt940/processors.py", "symbol": "mBank_set_tnr"}, "source_id": "csn:WoLpH/mt940:mt940/processors.py:mBank_set_tnr#7414"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/WoLpH/mt940/blob/fd71c88f7ec8372f4194f831f4c29c6f9bd9d0eb/mt940/processors.py#L73-L87", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:WoLpH-mt940:mBank_set_tnr-a2ae2bf062", "question": "mBank Collect states TNR in transaction details as unique id for\n transactions, that may be used to identify the same transactions in\n different statement files eg. partial mt942 and full mt940\n Information about tnr uniqueness has been obtained from mBank support,\n it lacks in mt940 mBank specification.", "scope": {"holder": "WoLpH-mt940", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Yaco-Sistemas/django-transmeta:transmeta/management/commands/sync_transmeta_db.py:Command.get_db_change_languages#14851", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "transmeta/management/commands/sync_transmeta_db.py", "symbol": "Command.get_db_change_languages"}, "source_id": "csn:Yaco-Sistemas/django-transmeta:transmeta/management/commands/sync_transmeta_db.py:Command.get_db_change_languages#14851"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Yaco-Sistemas/django-transmeta/blob/de070aae27770df046b4ba995f01f654db7ed1a2/transmeta/management/commands/sync_transmeta_db.py#L134-L145", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Yaco-Sistemas-django-transmeta:Command.get_db_change_languages-7733e38e30", "question": "get only db changes fields", "scope": {"holder": "Yaco-Sistemas-django-transmeta", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Yaco-Sistemas/django-transmeta:transmeta/management/commands/sync_transmeta_db.py:Command.handle#14850", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "transmeta/management/commands/sync_transmeta_db.py", "symbol": "Command.handle"}, "source_id": "csn:Yaco-Sistemas/django-transmeta:transmeta/management/commands/sync_transmeta_db.py:Command.handle#14850"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Yaco-Sistemas/django-transmeta/blob/de070aae27770df046b4ba995f01f654db7ed1a2/transmeta/management/commands/sync_transmeta_db.py#L63-L117", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Yaco-Sistemas-django-transmeta:Command.handle-87bdf7fec7", "question": "command execution", "scope": {"holder": "Yaco-Sistemas-django-transmeta", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Yelp/py_zipkin:py_zipkin/encoding/_helpers.py:Span.build_v1_span#7374", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "py_zipkin/encoding/_helpers.py", "symbol": "Span.build_v1_span"}, "source_id": "csn:Yelp/py_zipkin:py_zipkin/encoding/_helpers.py:Span.build_v1_span#7374"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Yelp/py_zipkin/blob/0944d9a3fb1f1798dbb276694aeed99f2b4283ba/py_zipkin/encoding/_helpers.py#L117-L153", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Yelp-py_zipkin:Span.build_v1_span-78a1303568", "question": "Builds and returns a V1 Span.\n\n :return: newly generated _V1Span\n :rtype: _V1Span", "scope": {"holder": "Yelp-py_zipkin", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Yelp/py_zipkin:py_zipkin/encoding/_helpers.py:copy_endpoint_with_new_service_name#7373", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "py_zipkin/encoding/_helpers.py", "symbol": "copy_endpoint_with_new_service_name"}, "source_id": "csn:Yelp/py_zipkin:py_zipkin/encoding/_helpers.py:copy_endpoint_with_new_service_name#7373"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Yelp/py_zipkin/blob/0944d9a3fb1f1798dbb276694aeed99f2b4283ba/py_zipkin/encoding/_helpers.py#L206-L220", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Yelp-py_zipkin:copy_endpoint_with_new_service_name-9213824200", "question": "Creates a copy of a given endpoint with a new service name.\n\n :param endpoint: existing Endpoint object\n :type endpoint: Endpoint\n :param new_service_name: new service name\n :type new_service_name: str\n :returns: zipkin new Endpoint object", "scope": {"holder": "Yelp-py_zipkin", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:Yubico/python-u2flib-host:u2flib_host/device.py:U2FDevice.get_supported_versions#12970", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "u2flib_host/device.py", "symbol": "U2FDevice.get_supported_versions"}, "source_id": "csn:Yubico/python-u2flib-host:u2flib_host/device.py:U2FDevice.get_supported_versions#12970"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Yubico/python-u2flib-host/blob/eadc4dbf3bf516e74ea00d2e5690742a535834cb/u2flib_host/device.py#L69-L80", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Yubico-python-u2flib-host:U2FDevice.get_supported_versions-729e2aa275", "question": "Gets a list of supported U2F versions from the device.", "scope": {"holder": "Yubico-python-u2flib-host", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:Yubico/python-u2flib-host:u2flib_host/u2f_v2.py:authenticate#12974", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "u2flib_host/u2f_v2.py", "symbol": "authenticate"}, "source_id": "csn:Yubico/python-u2flib-host:u2flib_host/u2f_v2.py:authenticate#12974"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/Yubico/python-u2flib-host/blob/eadc4dbf3bf516e74ea00d2e5690742a535834cb/u2flib_host/u2f_v2.py#L81-L126", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:Yubico-python-u2flib-host:authenticate-f3e9a132e0", "question": "Signs an authentication challenge\n\n data = {\n 'version': \"U2F_V2\",\n 'challenge': websafe_encode(self.challenge),\n 'appId': self.binding.app_id,\n 'keyHandle': websafe_encode(self.binding.key_handle)\n }", "scope": {"holder": "Yubico-python-u2flib-host", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ZELLMECHANIK-DRESDEN/fcswrite:examples/tdms2fcs.py:add_deformation#12908", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "examples/tdms2fcs.py", "symbol": "add_deformation"}, "source_id": "csn:ZELLMECHANIK-DRESDEN/fcswrite:examples/tdms2fcs.py:add_deformation#12908"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ZELLMECHANIK-DRESDEN/fcswrite/blob/5584983aa1eb927660183252039e73285c0724b3/examples/tdms2fcs.py#L32-L44", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ZELLMECHANIK-DRESDEN-fcswrite:add_deformation-76d0555270", "question": "From circularity, compute the deformation\n\n This method is useful for RT-DC data sets that contain\n the circularity but not the deformation.", "scope": {"holder": "ZELLMECHANIK-DRESDEN-fcswrite", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ZELLMECHANIK-DRESDEN/fcswrite:examples/tdms2fcs.py:tdms2fcs#12909", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "examples/tdms2fcs.py", "symbol": "tdms2fcs"}, "source_id": "csn:ZELLMECHANIK-DRESDEN/fcswrite:examples/tdms2fcs.py:tdms2fcs#12909"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ZELLMECHANIK-DRESDEN/fcswrite/blob/5584983aa1eb927660183252039e73285c0724b3/examples/tdms2fcs.py#L47-L54", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ZELLMECHANIK-DRESDEN-fcswrite:tdms2fcs-959f8a5053", "question": "Creates an fcs file for a given tdms file", "scope": {"holder": "ZELLMECHANIK-DRESDEN-fcswrite", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:a-tal/nagaram:nagaram/anagrams.py:_letter_map#19569", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "nagaram/anagrams.py", "symbol": "_letter_map"}, "source_id": "csn:a-tal/nagaram:nagaram/anagrams.py:_letter_map#19569"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/a-tal/nagaram/blob/2edcb0ef8cb569ebd1c398be826472b4831d6110/nagaram/anagrams.py#L7-L23", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:a-tal-nagaram:_letter_map-10173fda89", "question": "Creates a map of letter use in a word.\n\n Args:\n word: a string to create a letter map from\n\n Returns:\n a dictionary of {letter: integer count of letter in word}", "scope": {"holder": "a-tal-nagaram", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:a-tal/nagaram:nagaram/cmdline.py:main#19675", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "nagaram/cmdline.py", "symbol": "main"}, "source_id": "csn:a-tal/nagaram:nagaram/cmdline.py:main#19675"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/a-tal/nagaram/blob/2edcb0ef8cb569ebd1c398be826472b4831d6110/nagaram/cmdline.py#L148-L160", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:a-tal-nagaram:main-f72fd797b0", "question": "Main command line entry point.", "scope": {"holder": "a-tal-nagaram", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:abau171/highfive:highfive/jobs.py:JobManager._distribute_jobs#19322", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "highfive/jobs.py", "symbol": "JobManager._distribute_jobs"}, "source_id": "csn:abau171/highfive:highfive/jobs.py:JobManager._distribute_jobs#19322"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/abau171/highfive/blob/07b3829331072035ab100d1d66deca3e8f3f372a/highfive/jobs.py#L359-L370", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:abau171-highfive:JobManager._distribute_jobs-6924ada722", "question": "Distributes jobs from the active job set to any waiting get_job\n callbacks.", "scope": {"holder": "abau171-highfive", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:abau171/highfive:highfive/master.py:Master.run#19311", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "highfive/master.py", "symbol": "Master.run"}, "source_id": "csn:abau171/highfive:highfive/master.py:Master.run#19311"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/abau171/highfive/blob/07b3829331072035ab100d1d66deca3e8f3f372a/highfive/master.py#L183-L191", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:abau171-highfive:Master.run-1e82ee1150", "question": "Runs a job set which consists of the jobs in an iterable job list.", "scope": {"holder": "abau171-highfive", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:accraze/pymtranslate:pymtranslate/translator.py:Translator.convertArgsToTokens#16173", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pymtranslate/translator.py", "symbol": "Translator.convertArgsToTokens"}, "source_id": "csn:accraze/pymtranslate:pymtranslate/translator.py:Translator.convertArgsToTokens#16173"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/accraze/pymtranslate/blob/b18f9d7e8ef1583c988e8beb6c3304d362a4d979/pymtranslate/translator.py#L44-L62", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:accraze-pymtranslate:Translator.convertArgsToTokens-018e3c04b4", "question": "this converts the readin lines from\n sys to useable format, returns list\n of token and dict of tokens", "scope": {"holder": "accraze-pymtranslate", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:accraze/pymtranslate:pymtranslate/translator.py:Translator.iterateEM#16175", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pymtranslate/translator.py", "symbol": "Translator.iterateEM"}, "source_id": "csn:accraze/pymtranslate:pymtranslate/translator.py:Translator.iterateEM#16175"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/accraze/pymtranslate/blob/b18f9d7e8ef1583c988e8beb6c3304d362a4d979/pymtranslate/translator.py#L103-L179", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:accraze-pymtranslate:Translator.iterateEM-584bce293e", "question": "Iterate through all transmissions of english to\n foreign words. keep count of repeated occurences\n do until convergence\n set count(e|f) to 0 for all e,f\n set total(f) to 0 for all f\n for all sentence pairs (e_s,f_s)\n set total_s(e) = 0 for all e\n for all words e in e_s\n for all words f in f_s\n total_s(e) += t(e|f)\n for all words e in e_s\n for all words f in f_s\n count(e|f) += t(e|f) / total_s(e)\n total(f) += t(e|f) / total_s(e)\n for all f\n for all e\n t(e|f) = count(e|f) / total(f)", "scope": {"holder": "accraze-pymtranslate", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:adafruit/Adafruit_Python_PureIO:Adafruit_PureIO/smbus.py:SMBus.process_call#14104", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "Adafruit_PureIO/smbus.py", "symbol": "SMBus.process_call"}, "source_id": "csn:adafruit/Adafruit_Python_PureIO:Adafruit_PureIO/smbus.py:SMBus.process_call#14104"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/adafruit/Adafruit_Python_PureIO/blob/6f4976d91c52d70b67b28bba75a429b5328a52c1/Adafruit_PureIO/smbus.py#L296-L314", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:adafruit-Adafruit_Python_PureIO:SMBus.process_call-edd65038ad", "question": "Perform a smbus process call by writing a word (2 byte) value to\n the specified register of the device, and then reading a word of response\n data (which is returned).", "scope": {"holder": "adafruit-Adafruit_Python_PureIO", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:adafruit/Adafruit_Python_PureIO:Adafruit_PureIO/smbus.py:SMBus.write_byte#14098", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "Adafruit_PureIO/smbus.py", "symbol": "SMBus.write_byte"}, "source_id": "csn:adafruit/Adafruit_Python_PureIO:Adafruit_PureIO/smbus.py:SMBus.write_byte#14098"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/adafruit/Adafruit_Python_PureIO/blob/6f4976d91c52d70b67b28bba75a429b5328a52c1/Adafruit_PureIO/smbus.py#L232-L238", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:adafruit-Adafruit_Python_PureIO:SMBus.write_byte-185ed65c9c", "question": "Write a single byte to the specified device.", "scope": {"holder": "adafruit-Adafruit_Python_PureIO", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:addok/addok:addok/helpers/__init__.py:ChunkedPool.imap_unordered#9483", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "addok/helpers/__init__.py", "symbol": "ChunkedPool.imap_unordered"}, "source_id": "csn:addok/addok:addok/helpers/__init__.py:ChunkedPool.imap_unordered#9483"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/addok/addok/blob/46a270d76ec778d2b445c2be753e5c6ba070a9b2/addok/helpers/__init__.py#L144-L164", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:addok-addok:ChunkedPool.imap_unordered-08486e4b80", "question": "Customized version of imap_unordered.\n\n Directly send chunks to func, instead of iterating in each process and\n sending one by one.\n\n Original:\n https://hg.python.org/cpython/file/tip/Lib/multiprocessing/pool.py#l271\n\n Other tried options:\n - map_async: makes a list(iterable), so it loads all the data for each\n process into RAM\n - apply_async: needs manual chunking", "scope": {"holder": "addok-addok", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:addok/addok:addok/shell.py:Cmd.do_INTERSECT#9490", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "addok/shell.py", "symbol": "Cmd.do_INTERSECT"}, "source_id": "csn:addok/addok:addok/shell.py:Cmd.do_INTERSECT#9490"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/addok/addok/blob/46a270d76ec778d2b445c2be753e5c6ba070a9b2/addok/shell.py#L218-L234", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:addok-addok:Cmd.do_INTERSECT-18e8b24e97", "question": "Do a raw intersect between tokens (default limit 100).\n INTERSECT rue des lilas [LIMIT 100]", "scope": {"holder": "addok-addok", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:agabrown/PyGaia:pygaia/astrometry/coordinates.py:CoordinateTransformation._getJacobian#11912", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pygaia/astrometry/coordinates.py", "symbol": "CoordinateTransformation._getJacobian"}, "source_id": "csn:agabrown/PyGaia:pygaia/astrometry/coordinates.py:CoordinateTransformation._getJacobian#11912"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/agabrown/PyGaia/blob/ae972b0622a15f713ffae471f925eac25ccdae47/pygaia/astrometry/coordinates.py#L298-L342", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:agabrown-PyGaia:CoordinateTransformation._getJacobian-5ee61af87e", "question": "Calculates the Jacobian for the transformation of the position errors and proper motion errors\n between coordinate systems. This Jacobian is also the rotation matrix for the transformation of\n proper motions. See section 1.5.3 of the Hipparcos Explanatory Volume 1 (equation 1.5.20). This\n matrix has the following form:\n\n | c s |\n J = | |\n | -s c |\n\n Parameters\n ----------\n\n phi - The longitude-like angle of the position of the source (radians).\n theta - The latitude-like angle of the position of the source (radians).\n\n Returns\n -------\n\n c, s - The Jacobian matrix elements c and s corresponding to (phi, theta) and the currently\n desired coordinate system transformation.", "scope": {"holder": "agabrown-PyGaia", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:agabrown/PyGaia:examples/radialVelocityHorizons.py:makePlot#11916", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "examples/radialVelocityHorizons.py", "symbol": "makePlot"}, "source_id": "csn:agabrown/PyGaia:examples/radialVelocityHorizons.py:makePlot#11916"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/agabrown/PyGaia/blob/ae972b0622a15f713ffae471f925eac25ccdae47/examples/radialVelocityHorizons.py#L35-L129", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:agabrown-PyGaia:makePlot-2c382b0b29", "question": "Make the plot with parallax horizons. The plot shows V-band magnitude vs distance for a number of\n spectral types and over the range 5.7 `self.size`.", "scope": {"holder": "alghafli-fileslice", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:algofairness/BlackBoxAuditing:python2_source/BlackBoxAuditing/measurements.py:BCR#9437", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "python2_source/BlackBoxAuditing/measurements.py", "symbol": "BCR"}, "source_id": "csn:algofairness/BlackBoxAuditing:python2_source/BlackBoxAuditing/measurements.py:BCR#9437"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/algofairness/BlackBoxAuditing/blob/b06c4faed5591cd7088475b2a203127bc5820483/python2_source/BlackBoxAuditing/measurements.py#L14-L30", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:algofairness-BlackBoxAuditing:BCR-85fc417701", "question": "Given a confusion matrix, returns Balanced Classification Rate.\n BCR is (1 - Balanced Error Rate).\n BER Definition: http://research.ics.aalto.fi/events/eyechallenge2005/evaluation.shtml", "scope": {"holder": "algofairness-BlackBoxAuditing", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:algofairness/BlackBoxAuditing:BlackBoxAuditing/repairers/calculators.py:get_median#9438", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "BlackBoxAuditing/repairers/calculators.py", "symbol": "get_median"}, "source_id": "csn:algofairness/BlackBoxAuditing:BlackBoxAuditing/repairers/calculators.py:get_median#9438"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/algofairness/BlackBoxAuditing/blob/b06c4faed5591cd7088475b2a203127bc5820483/BlackBoxAuditing/repairers/calculators.py#L3-L24", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:algofairness-BlackBoxAuditing:get_median-9cfaeedf31", "question": "Given an unsorted list of numeric values, return median value (as a float).\n Note that in the case of even-length lists of values, we apply the value to\n the left of the center to be the median (such that the median can only be\n a value from the list of values).\n Eg: get_median([1,2,3,4]) == 2, not 2.5.", "scope": {"holder": "algofairness-BlackBoxAuditing", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:alimuldal/PyFNND:pyfnnd/demo.py:make_fake_movie#17655", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyfnnd/demo.py", "symbol": "make_fake_movie"}, "source_id": "csn:alimuldal/PyFNND:pyfnnd/demo.py:make_fake_movie#17655"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/alimuldal/PyFNND/blob/3cbe0622a385f5206837bfd944d781aa7b1649ea/pyfnnd/demo.py#L8-L73", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:alimuldal-PyFNND:make_fake_movie-368566c8b6", "question": "Generate 2D fake fluorescence movie\n\n Arguments:\n ---------------------------------------------------------------------------\n nframes: number of timebins to simulate\n mask_shape: tuple (nrows, ncols), shape of a single movie frame\n mask_center: tuple (x, y), pixel coords of cell center\n bg_intensity: scalar, amplitude of (static) baseline fluorescence\n mask_sigma: scalar, standard deviation of Gaussian mask\n dt: timestep (s)\n rate: mean spike rate (Hz)\n tau: time constant of decay in calcium concentration (s)\n sigma: SD of additive noise on fluorescence\n seed: Seed for RNG\n\n Returns:\n ---------------------------------------------------------------------------\n F: fluorescence [npixels, nframes]\n c: calcium concentration [nframes,]\n n: spike train [nframes,]\n theta: tuple of true model parameters:\n (sigma, alpha, beta, lambda, gamma)", "scope": {"holder": "alimuldal-PyFNND", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:alimuldal/PyFNND:pyfnnd/utils.py:s2h#17660", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "pyfnnd/utils.py", "symbol": "s2h"}, "source_id": "csn:alimuldal/PyFNND:pyfnnd/utils.py:s2h#17660"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/alimuldal/PyFNND/blob/3cbe0622a385f5206837bfd944d781aa7b1649ea/pyfnnd/utils.py#L24-L34", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:alimuldal-PyFNND:s2h-f4ca2d684a", "question": "convert seconds to a pretty \"d hh:mm:ss.s\" format", "scope": {"holder": "alimuldal-PyFNND", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:allanlei/django-argparse-command:argcmd/management/base.py:BaseCommand.create_parser#20599", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "argcmd/management/base.py", "symbol": "BaseCommand.create_parser"}, "source_id": "csn:allanlei/django-argparse-command:argcmd/management/base.py:BaseCommand.create_parser#20599"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/allanlei/django-argparse-command/blob/27ea77e1dd0cf2f0567223735762a5ebd14fdaef/argcmd/management/base.py#L46-L61", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:allanlei-django-argparse-command:BaseCommand.create_parser-e2a6343a44", "question": "Create and return the ``ArgumentParser`` which will be used to\n parse the arguments to this command.", "scope": {"holder": "allanlei-django-argparse-command", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:alpacahq/alpaca-trade-api-python:alpaca_trade_api/rest.py:REST._one_request#4543", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "alpaca_trade_api/rest.py", "symbol": "REST._one_request"}, "source_id": "csn:alpacahq/alpaca-trade-api-python:alpaca_trade_api/rest.py:REST._one_request#4543"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/alpacahq/alpaca-trade-api-python/blob/9c9dea3b4a37c909f88391b202e86ff356a8b4d7/alpaca_trade_api/rest.py#L111-L134", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:alpacahq-alpaca-trade-api-python:REST._one_request-589595b454", "question": "Perform one request, possibly raising RetryException in the case\n the response is 429. Otherwise, if error text contain \"code\" string,\n then it decodes to json object and returns APIError.\n Returns the body json in the 200 status.", "scope": {"holder": "alpacahq-alpaca-trade-api-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:alpacahq/alpaca-trade-api-python:alpaca_trade_api/rest.py:REST.list_orders#4544", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "alpaca_trade_api/rest.py", "symbol": "REST.list_orders"}, "source_id": "csn:alpacahq/alpaca-trade-api-python:alpaca_trade_api/rest.py:REST.list_orders#4544"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/alpacahq/alpaca-trade-api-python/blob/9c9dea3b4a37c909f88391b202e86ff356a8b4d7/alpaca_trade_api/rest.py#L154-L173", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:alpacahq-alpaca-trade-api-python:REST.list_orders-e106d6c3de", "question": "Get a list of orders\n https://docs.alpaca.markets/web-api/orders/#get-a-list-of-orders", "scope": {"holder": "alpacahq-alpaca-trade-api-python", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:ambitioninc/python-logentries-api:logentries_api/resources.py:Alerts.get#18163", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "logentries_api/resources.py", "symbol": "Alerts.get"}, "source_id": "csn:ambitioninc/python-logentries-api:logentries_api/resources.py:Alerts.get#18163"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ambitioninc/python-logentries-api/blob/77ff1a7a2995d7ea2725b74e34c0f880f4ee23bc/logentries_api/resources.py#L560-L591", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ambitioninc-python-logentries-api:Alerts.get-a73e1ec293", "question": "Get alerts that match the alert type and args.\n\n :param alert_type: The type of the alert. Must be one of 'pagerduty',\n 'mailto', 'webhook', 'slack', or 'hipchat'\n :type alert_type: str\n\n :param alert_args: The args for the alert. The provided args must be a\n subset of the actual alert args. If no args are provided, all\n alerts matching the ``alert_type`` are returned. For example:\n ``.get('mailto', alert_args={'direct': 'me@mydomain.com'})`` or\n ``.get('slack', {'url': 'https://hooks.slack.com/services...'})``\n\n :return: A list of matching alerts. An empty list is returned if there\n are not any matches\n :rtype: list of dict\n\n :raises: This will raise a\n :class:`ServerException`\n if there is an error from Logentries", "scope": {"holder": "ambitioninc-python-logentries-api", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:ambitioninc/python-logentries-api:logentries_api/resources.py:Labels.update#18153", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "logentries_api/resources.py", "symbol": "Labels.update"}, "source_id": "csn:ambitioninc/python-logentries-api:logentries_api/resources.py:Labels.update#18153"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ambitioninc/python-logentries-api/blob/77ff1a7a2995d7ea2725b74e34c0f880f4ee23bc/logentries_api/resources.py#L151-L192", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ambitioninc-python-logentries-api:Labels.update-84d27ebdcf", "question": "Update a Label\n\n :param label: The data to update. Must include keys:\n\n * id (str)\n * appearance (dict)\n * description (str)\n * name (str)\n * title (str)\n :type label: dict\n\n Example:\n\n .. code-block:: python\n\n Labels().update(\n label={\n 'id': 'd9d4596e-49e4-4135-b3b3-847f9e7c1f43',\n 'appearance': {'color': '278abe'},\n 'name': 'My Sandbox',\n 'description': 'My Sandbox',\n 'title': 'My Sandbox',\n }\n )\n\n :return:\n :rtype: dict", "scope": {"holder": "ambitioninc-python-logentries-api", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:amcat/progressmonitor:progressmonitor/__init__.py:ProgressMonitor.begin#20822", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "progressmonitor/__init__.py", "symbol": "ProgressMonitor.begin"}, "source_id": "csn:amcat/progressmonitor:progressmonitor/__init__.py:ProgressMonitor.begin#20822"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amcat/progressmonitor/blob/d4cabebc95bfd1447120f601c094b20bee954285/progressmonitor/__init__.py#L72-L77", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amcat-progressmonitor:ProgressMonitor.begin-cfa7b1e9b1", "question": "Call before starting work on a monitor, specifying name and amount of work", "scope": {"holder": "amcat-progressmonitor", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:amcat/progressmonitor:progressmonitor/__init__.py:ProgressMonitor.task#20823", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "progressmonitor/__init__.py", "symbol": "ProgressMonitor.task"}, "source_id": "csn:amcat/progressmonitor:progressmonitor/__init__.py:ProgressMonitor.task#20823"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amcat/progressmonitor/blob/d4cabebc95bfd1447120f601c094b20bee954285/progressmonitor/__init__.py#L80-L86", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amcat-progressmonitor:ProgressMonitor.task-878e65b887", "question": "Wrap code into a begin and end call on this monitor", "scope": {"holder": "amcat-progressmonitor", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:amelchio/eternalegypt:eternalegypt/eternalegypt.py:autologin#14823", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "eternalegypt/eternalegypt.py", "symbol": "autologin"}, "source_id": "csn:amelchio/eternalegypt:eternalegypt/eternalegypt.py:autologin#14823"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amelchio/eternalegypt/blob/895e0b235ceaf7f61458c620237c3ad397780e98/eternalegypt/eternalegypt.py#L52-L71", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amelchio-eternalegypt:autologin-8452358fe5", "question": "Decorator that will try to login and redo an action before failing.", "scope": {"holder": "amelchio-eternalegypt", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:amelchio/eternalegypt:examples/status.py:get_information#14826", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "examples/status.py", "symbol": "get_information"}, "source_id": "csn:amelchio/eternalegypt:examples/status.py:get_information#14826"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amelchio/eternalegypt/blob/895e0b235ceaf7f61458c620237c3ad397780e98/examples/status.py#L11-L41", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amelchio-eternalegypt:get_information-b336a13310", "question": "Example of printing the current upstream.", "scope": {"holder": "amelchio-eternalegypt", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:amorison/loam:loam/cli.py:CLIManager._add_options_to_parser#19295", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "loam/cli.py", "symbol": "CLIManager._add_options_to_parser"}, "source_id": "csn:amorison/loam:loam/cli.py:CLIManager._add_options_to_parser#19295"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amorison/loam/blob/a566c943a75e068a4510099331a1ddfe5bbbdd94/loam/cli.py#L141-L154", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amorison-loam:CLIManager._add_options_to_parser-7a1b874349", "question": "Add options to a parser.", "scope": {"holder": "amorison-loam", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:amorison/loam:loam/manager.py:ConfigurationManager.read_configs_#19290", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "loam/manager.py", "symbol": "ConfigurationManager.read_configs_"}, "source_id": "csn:amorison/loam:loam/manager.py:ConfigurationManager.read_configs_#19290"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amorison/loam/blob/a566c943a75e068a4510099331a1ddfe5bbbdd94/loam/manager.py#L366-L389", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amorison-loam:ConfigurationManager.read_configs_-26ed94d364", "question": "Read config files and set config values accordingly.\n\n Returns:\n (dict, list, list): respectively content of files, list of\n missing/empty files and list of files for which a parsing error\n arised.", "scope": {"holder": "amorison-loam", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:amyth/django-instapush:instapush/models/base.py:GCMDevice.send_message#17148", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "instapush/models/base.py", "symbol": "GCMDevice.send_message"}, "source_id": "csn:amyth/django-instapush:instapush/models/base.py:GCMDevice.send_message#17148"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amyth/django-instapush/blob/f8643a2e342fc73a16c95dff79c3daac8ce4b034/instapush/models/base.py#L63-L74", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amyth-django-instapush:GCMDevice.send_message-3d6e50b667", "question": "Sends a push notification to this device via GCM", "scope": {"holder": "amyth-django-instapush", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:amyth/django-instapush:instapush/libs/apns.py:apns_fetch_inactive_ids#17150", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "instapush/libs/apns.py", "symbol": "apns_fetch_inactive_ids"}, "source_id": "csn:amyth/django-instapush:instapush/libs/apns.py:apns_fetch_inactive_ids#17150"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/amyth/django-instapush/blob/f8643a2e342fc73a16c95dff79c3daac8ce4b034/instapush/libs/apns.py#L245-L256", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:amyth-django-instapush:apns_fetch_inactive_ids-5129563f34", "question": "Queries the APNS server for id's that are no longer active since\n the last fetch", "scope": {"holder": "amyth-django-instapush", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:analytehealth/django-analytics:djanalytics/reports/utils.py:average_duration#15386", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "djanalytics/reports/utils.py", "symbol": "average_duration"}, "source_id": "csn:analytehealth/django-analytics:djanalytics/reports/utils.py:average_duration#15386"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/analytehealth/django-analytics/blob/7782d3f81249dcb1b266afb0cb1e90000108c74d/djanalytics/reports/utils.py#L12-L19", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:analytehealth-django-analytics:average_duration-39aa544068", "question": "Method to calculate and format an average duration safely", "scope": {"holder": "analytehealth-django-analytics", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:andrefsp/pyflot:flot/__init__.py:Graph._get_axis_mode#18903", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "flot/__init__.py", "symbol": "Graph._get_axis_mode"}, "source_id": "csn:andrefsp/pyflot:flot/__init__.py:Graph._get_axis_mode#18903"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/andrefsp/pyflot/blob/f2dde10709aeed39074fcce8172184b5cd8bfd66/flot/__init__.py#L266-L270", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:andrefsp-pyflot:Graph._get_axis_mode-9f2273e610", "question": "will get the axis mode for the current series", "scope": {"holder": "andrefsp-pyflot", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:andrefsp/pyflot:flot/__init__.py:Series._set_data#18902", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "flot/__init__.py", "symbol": "Series._set_data"}, "source_id": "csn:andrefsp/pyflot:flot/__init__.py:Series._set_data#18902"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/andrefsp/pyflot/blob/f2dde10709aeed39074fcce8172184b5cd8bfd66/flot/__init__.py#L168-L186", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:andrefsp-pyflot:Series._set_data-5b2c4c6aa9", "question": "This method will be called to set Series data", "scope": {"holder": "andrefsp-pyflot", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:andreikop/python-ws-discovery:wsdiscovery/daemon.py:NetworkingThread.addSourceAddr#7517", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "wsdiscovery/daemon.py", "symbol": "NetworkingThread.addSourceAddr"}, "source_id": "csn:andreikop/python-ws-discovery:wsdiscovery/daemon.py:NetworkingThread.addSourceAddr#7517"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/andreikop/python-ws-discovery/blob/a7b852cf43115c6f986e509b1870d6963e76687f/wsdiscovery/daemon.py#L126-L135", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:andreikop-python-ws-discovery:NetworkingThread.addSourceAddr-0557345ca7", "question": "None means 'system default", "scope": {"holder": "andreikop-python-ws-discovery", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:andreikop/python-ws-discovery:wsdiscovery/daemon.py:WSDiscovery.setRemoteServiceHelloCallback#7519", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "wsdiscovery/daemon.py", "symbol": "WSDiscovery.setRemoteServiceHelloCallback"}, "source_id": "csn:andreikop/python-ws-discovery:wsdiscovery/daemon.py:WSDiscovery.setRemoteServiceHelloCallback#7519"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/andreikop/python-ws-discovery/blob/a7b852cf43115c6f986e509b1870d6963e76687f/wsdiscovery/daemon.py#L291-L303", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:andreikop-python-ws-discovery:WSDiscovery.setRemoteServiceHelloCallback-f84704e6fc", "question": "Set callback, which will be called when new service appeared online\n and sent Hi message\n\n typesFilter and scopesFilter might be list of types and scopes.\n If filter is set, callback is called only for Hello messages,\n which match filter\n\n Set None to disable callback", "scope": {"holder": "andreikop-python-ws-discovery", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:anentropic/dirtyebay:dirtyebay/utils.py:_default_ns_prefix#21865", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "dirtyebay/utils.py", "symbol": "_default_ns_prefix"}, "source_id": "csn:anentropic/dirtyebay:dirtyebay/utils.py:_default_ns_prefix#21865"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/anentropic/dirtyebay/blob/c3194932360445563cfe0c7e034e41f09358c7eb/dirtyebay/utils.py#L60-L90", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:anentropic-dirtyebay:_default_ns_prefix-4d1afb181a", "question": "XML doc may have several prefix:namespace_url pairs, can also specify\n a namespace_url as default, tags in that namespace don't need a prefix\n NOTE:\n we rely on default namespace also present in prefixed form, I'm not sure if\n this is an XML certainty or a quirk of the eBay WSDLs\n\n in our case the WSDL contains:\n \n 1.0.0\n \n ...but our query needs to give a prefix to the path of `Version` so we need\n to determine the default namespace of the doc, find the matching prefix and\n return it", "scope": {"holder": "anentropic-dirtyebay", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:anentropic/dirtyebay:dirtyebay/utils.py:version_from_schema#21864", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "dirtyebay/utils.py", "symbol": "version_from_schema"}, "source_id": "csn:anentropic/dirtyebay:dirtyebay/utils.py:version_from_schema#21864"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/anentropic/dirtyebay/blob/c3194932360445563cfe0c7e034e41f09358c7eb/dirtyebay/utils.py#L36-L57", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:anentropic-dirtyebay:version_from_schema-fadf23a2d9", "question": "returns: API version number \n raises: \n\n NOTE:\n relies on presence of comment tags in the XSD, which are currently present\n for both ebaySvc.xsd (TradingAPI) and ShoppingService.xsd (ShoppingAPI)", "scope": {"holder": "anentropic-dirtyebay", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:ankitmathur3193/song-cli:song/commands/SearchEngineParser/GoogleParser.py:GoogleParser.Parse#17683", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "song/commands/SearchEngineParser/GoogleParser.py", "symbol": "GoogleParser.Parse"}, "source_id": "csn:ankitmathur3193/song-cli:song/commands/SearchEngineParser/GoogleParser.py:GoogleParser.Parse#17683"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ankitmathur3193/song-cli/blob/ca8ccfe547e9d702313ff6d14e81ae4355989a67/song/commands/SearchEngineParser/GoogleParser.py#L33-L43", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ankitmathur3193-song-cli:GoogleParser.Parse-2b9e9ab31a", "question": "song_name is a list of strings\n\t\twebsite is a string\n\t\tIt will return the url from where music file needs to be downloaded", "scope": {"holder": "ankitmathur3193-song-cli", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ankitmathur3193/song-cli:song/commands/SearchEngineParser/GoogleParser.py:GoogleParser.parse_google#17682", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "song/commands/SearchEngineParser/GoogleParser.py", "symbol": "GoogleParser.parse_google"}, "source_id": "csn:ankitmathur3193/song-cli:song/commands/SearchEngineParser/GoogleParser.py:GoogleParser.parse_google#17682"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ankitmathur3193/song-cli/blob/ca8ccfe547e9d702313ff6d14e81ae4355989a67/song/commands/SearchEngineParser/GoogleParser.py#L20-L30", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ankitmathur3193-song-cli:GoogleParser.parse_google-a0ecf6cb30", "question": "It will parse google html response\n\t\t\tand return the first url", "scope": {"holder": "ankitmathur3193-song-cli", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:ansible/ansible-lint:lib/ansiblelint/utils.py:normalize_task_v2#3564", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "lib/ansiblelint/utils.py", "symbol": "normalize_task_v2"}, "source_id": "csn:ansible/ansible-lint:lib/ansiblelint/utils.py:normalize_task_v2#3564"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ansible/ansible-lint/blob/b4a8743794e592698c32e760bc774a85f8eebeb5/lib/ansiblelint/utils.py#L368-L416", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ansible-ansible-lint:normalize_task_v2-9eb0c41b8f", "question": "Ensures tasks have an action key and strings are converted to python objects", "scope": {"holder": "ansible-ansible-lint", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:ansible/ansible-lint:lib/ansiblelint/utils.py:parse_yaml_linenumbers#3565", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "lib/ansiblelint/utils.py", "symbol": "parse_yaml_linenumbers"}, "source_id": "csn:ansible/ansible-lint:lib/ansiblelint/utils.py:parse_yaml_linenumbers#3565"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/ansible/ansible-lint/blob/b4a8743794e592698c32e760bc774a85f8eebeb5/lib/ansiblelint/utils.py#L546-L585", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:ansible-ansible-lint:parse_yaml_linenumbers-344302e324", "question": "Parses yaml as ansible.utils.parse_yaml but with linenumbers.\n\n The line numbers are stored in each node's LINE_NUMBER_KEY key.", "scope": {"holder": "ansible-ansible-lint", "scenario": "code-search", "session": null}, "split": "dev", "unanswerable": false} +{"answer": "csn:antevens/listen:listen/signal_handler.py:SignalHandler._unreg_event#16617", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "listen/signal_handler.py", "symbol": "SignalHandler._unreg_event"}, "source_id": "csn:antevens/listen:listen/signal_handler.py:SignalHandler._unreg_event#16617"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/antevens/listen/blob/d3ddff8e7fbfb672c5bd7f6f4febeb5e921d8c67/listen/signal_handler.py#L190-L204", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:antevens-listen:SignalHandler._unreg_event-f112f63ce1", "question": "Tries to remove a registered event without triggering it", "scope": {"holder": "antevens-listen", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:antevens/listen:listen/signal_handler.py:SignalHandler.reg_on_abort#16619", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "listen/signal_handler.py", "symbol": "SignalHandler.reg_on_abort"}, "source_id": "csn:antevens/listen:listen/signal_handler.py:SignalHandler.reg_on_abort#16619"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/antevens/listen/blob/d3ddff8e7fbfb672c5bd7f6f4febeb5e921d8c67/listen/signal_handler.py#L240-L245", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:antevens-listen:SignalHandler.reg_on_abort-a58a5bab8c", "question": "Register a function/method to be called when execution is aborted", "scope": {"holder": "antevens-listen", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:apache/airflow:airflow/contrib/hooks/aws_athena_hook.py:AWSAthenaHook.check_query_status#651", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "airflow/contrib/hooks/aws_athena_hook.py", "symbol": "AWSAthenaHook.check_query_status"}, "source_id": "csn:apache/airflow:airflow/contrib/hooks/aws_athena_hook.py:AWSAthenaHook.check_query_status#651"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/apache/airflow/blob/b69c686ad8a0c89b9136bb4b31767257eb7b2597/airflow/contrib/hooks/aws_athena_hook.py#L74-L89", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:apache-airflow:AWSAthenaHook.check_query_status-d861e509da", "question": "Fetch the status of submitted athena query. Returns None or one of valid query states.\n\n :param query_execution_id: Id of submitted athena query\n :type query_execution_id: str\n :return: str", "scope": {"holder": "apache-airflow", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:apache/airflow:airflow/contrib/sensors/imap_attachment_sensor.py:ImapAttachmentSensor.poke#753", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "airflow/contrib/sensors/imap_attachment_sensor.py", "symbol": "ImapAttachmentSensor.poke"}, "source_id": "csn:apache/airflow:airflow/contrib/sensors/imap_attachment_sensor.py:ImapAttachmentSensor.poke#753"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/apache/airflow/blob/b69c686ad8a0c89b9136bb4b31767257eb7b2597/airflow/contrib/sensors/imap_attachment_sensor.py#L60-L76", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:apache-airflow:ImapAttachmentSensor.poke-f96f856db6", "question": "Pokes for a mail attachment on the mail server.\n\n :param context: The context that is being provided when poking.\n :type context: dict\n :return: True if attachment with the given name is present and False if not.\n :rtype: bool", "scope": {"holder": "apache-airflow", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:appier/h5sparse:h5sparse/h5sparse.py:Group.create_dataset#12251", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "h5sparse/h5sparse.py", "symbol": "Group.create_dataset"}, "source_id": "csn:appier/h5sparse:h5sparse/h5sparse.py:Group.create_dataset#12251"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/appier/h5sparse/blob/b12ee65c624207d3020c283ce561cfcc7ab71731/h5sparse/h5sparse.py#L46-L94", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:appier-h5sparse:Group.create_dataset-ae16f3f8f4", "question": "Create 3 datasets in a group to represent the sparse array.\n\n Parameters\n ----------\n sparse_format:", "scope": {"holder": "appier-h5sparse", "scenario": "code-search", "session": null}, "split": "test", "unanswerable": false} +{"answer": "csn:appknox/google-chartwrapper:GChartWrapper/GChart.py:Axes.label#19516", "answer_format": "code", "as_of": "2026-02-23T14:11:41+00:00", "gold": [{"region": {"path": "GChartWrapper/GChart.py", "symbol": "Axes.label"}, "source_id": "csn:appknox/google-chartwrapper:GChartWrapper/GChart.py:Axes.label#19516"}], "modality": "code", "origin": {"dataset": "code-search-net/code_search_net", "item": "url:https://github.com/appknox/google-chartwrapper/blob/3769aecbef6c83b6cd93ee72ece478ffe433ac57/GChartWrapper/GChart.py#L122-L131", "revision": "bd0cf261e357a3eb5c8fba490d23ec1a1cd59555"}, "query_id": "csn:appknox-google-chartwrapper:Axes.label-470acef192", "question": "Label each axes one at a time\n args are of the form