Duplicate from dzxr/chinese-classical-corpus

Co-authored-by: Zion L <dzxr@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,63 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.avro filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.lz4 filter=lfs diff=lfs merge=lfs -text
+*.mds filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+# Audio files - uncompressed
+*.pcm filter=lfs diff=lfs merge=lfs -text
+*.sam filter=lfs diff=lfs merge=lfs -text
+*.raw filter=lfs diff=lfs merge=lfs -text
+# Audio files - compressed
+*.aac filter=lfs diff=lfs merge=lfs -text
+*.flac filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.ogg filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text
+# Image files - uncompressed
+*.bmp filter=lfs diff=lfs merge=lfs -text
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+# Image files - compressed
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.jpeg filter=lfs diff=lfs merge=lfs -text
+*.webp filter=lfs diff=lfs merge=lfs -text
+# Video files - compressed
+*.mp4 filter=lfs diff=lfs merge=lfs -text
+*.webm filter=lfs diff=lfs merge=lfs -text
+corpus.jsonl filter=lfs diff=lfs merge=lfs -text
+punctuate.jsonl filter=lfs diff=lfs merge=lfs -text
+translate.jsonl filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,159 @@
+---
+license: cc0-1.0
+language:
+  - zh
+task_categories:
+  - text-generation
+  - translation
+tags:
+  - chinese-classical
+  - 古文
+  - 文言文
+  - instruction-tuning
+  - llm
+size_categories:
+  - 1M<n<10M
+configs:
+  - config_name: corpus
+    data_files:
+      - split: train
+        path: "corpus.jsonl"
+  - config_name: translate
+    data_files:
+      - split: train
+        path: "translate.jsonl"
+  - config_name: punctuate
+    data_files:
+      - split: train
+        path: "punctuate.jsonl"
+---
+
+# Chinese Classical Corpus
+
+> 🔗 **Source code & build scripts**: [github.com/zi6me/chinese-classical-corpus](https://github.com/zi6me/chinese-classical-corpus) — full extraction pipeline, 14 Python scripts, validation suite.
+
+中国古典文献结构化语料集，含完整十三经 + 说文解字 + 资治通鉴 + 二十四史前 15 部，以及 197 万条古译今/今译古/断句指令对。
+
+**全部 CC0 公有领域，可商用、可改用、无附加限制。**
+
+## Quick Start
+
+```python
+from datasets import load_dataset
+
+# 源语料 (12,005 条章节级记录, 17.2M 字)
+corpus = load_dataset("dzxr/chinese-classical-corpus", "corpus", split="train")
+
+# 古译今 / 今译古 双向指令数据 (1,924,378 条)
+translate = load_dataset("dzxr/chinese-classical-corpus", "translate", split="train")
+
+# 断句加标点指令 (46,546 条)
+punctuate = load_dataset("dzxr/chinese-classical-corpus", "punctuate", split="train")
+
+print(corpus[0])
+# {'id': 'shuowen#1', 'source': '说文解字', 'author': '许慎', 'era': '汉',
+#  'category': '字书', 'char': '一', 'radical': '一部', ...}
+```
+
+## Configs
+
+### `corpus` — 源语料
+
+12,005 条章节/篇/卷级别的清洁文本，覆盖 30 部典籍：
+
+| 类别 | 文献 | 备注 |
+|------|------|------|
+| 字书 | 说文解字 (9,831 字头) | 21% 字头修复至 3.6% (CJK Ext B-G 通过 [shuowenjiezi/shuowen](https://github.com/shuowenjiezi/shuowen) 交叉补全) |
+| 经 (十三经) | 论语、孟子、大学、中庸、诗经、尚书、礼记、周易、春秋左传/公羊传/穀梁传、孝经、尔雅 | 完整 |
+| 史 | 史记、汉书、后汉书、三国志、晋书、宋书、南齐书、梁书、陈书、魏书、北齐书、周书、南史、北史、隋书 (二十四史前 15 部) | |
+| 编年 | 资治通鉴 294 卷 | 完整 |
+
+**通用字段**：`id` `source` `author` `era` `category` `content`
+**类型特定**：字书加 `char/radical/pinyin/fanqie`，经类加 `chapter/section`，史类加 `volume`
+
+### `translate` — 古译今 / 今译古
+
+1,924,378 条指令记录，覆盖 97 部典籍。来源：[NiuTrans/Classical-Modern](https://github.com/NiuTrans/Classical-Modern) (MIT) 双语数据。
+
+```json
+{
+  "id": "instruct#1",
+  "task": "c2m",
+  "instruction": "将下列古文翻译成现代汉语：",
+  "input": "子曰：学而时习之，不亦说乎？",
+  "output": "孔子说：学了知识然后按一定的时间复习它，不也是很愉快吗？",
+  "source": "论语·学而篇",
+  "category": "经"
+}
+```
+
+- `task`: `c2m` (古→今) 或 `m2c` (今→古)
+- 6 种 c2m 指令 + 4 种 m2c 指令模板，轮换使用
+- 798 条记录有 `_has_box: true` flag — 含古籍散佚 □ 占位符（参见下方说明）
+
+### `punctuate` — 断句加标点
+
+46,546 条记录，从 `corpus` 中抽取章节段，去掉标点作为输入，原文作为输出。覆盖 14 部正史 + 经传。
+
+```json
+{
+  "id": "punct#1",
+  "task": "punctuate",
+  "instruction": "为下列古文添加标点：",
+  "input": "夫天地者万物之逆旅也光阴者百代之过客也",
+  "output": "夫天地者，万物之逆旅也；光阴者，百代之过客也。",
+  "source": "晋书",
+  "category": "史"
+}
+```
+
+## 关于 □ 字符
+
+NiuTrans 源中 1,266 条记录含 □（占指令数据 0.06%），分两类：
+
+1. **可恢复（678 条已修复）** — 通过 chtxt 繁体版交叉补全。例：`营昭阳殿，□令监造` → `营昭阳殿，𠡠令监造`（𠡠 是 CJK Ext B U+2086D 的"敕"异体字）
+2. **不可恢复（798 条加 `_has_box: true`）** — 古籍散佚（逸周书）、出土文献残损（孙膑兵法竹简）、帛书重建（黄帝四经）、礼乐符号（礼记节奏记号）
+
+训练时按需过滤：`ds.filter(lambda x: not x.get("_has_box", False))`
+
+## 数据来源
+
+- [殆知阁古代文献](https://github.com/garychowcmu/daizhigev20) — 主要语料源
+- [chinese-poetry](https://github.com/chinese-poetry/chinese-poetry) — 诗经 + 四书
+- [zh.wikisource.org](https://zh.wikisource.org) — 穀梁传
+- [ctext.org](https://ctext.org) — 尚书/周易缺漏补全
+- [shuowenjiezi/shuowen](https://github.com/shuowenjiezi/shuowen) — 说文 □ 字修复
+- [chtxt](https://github.com/JasonWade001/chtxt) — 二十四史清洁版繁体（9 部 繁→简）
+- [NiuTrans/Classical-Modern](https://github.com/NiuTrans/Classical-Modern) (MIT) — 古译今对齐 + 南史 76 卷
+
+## 已知数据问题
+
+- **说文 356 字仍为 □** (3.6%)：跨参考源也无法消歧
+- **资治通鉴 卷 258**：源文中作者属字误为 `寀` (cǎi)，实为 `宋` (sòng)
+- **南史 76/80 卷**：NiuTrans 上游缺最末 4 卷
+- **生僻字 ASCII 化**：极个别 CJK Ext G 字符在转换中可能丢失
+
+## 路线图
+
+- v1.3 — 加二十四史下 9 部（旧/新唐书、旧/新五代史、宋史、辽金元明史 + 清史稿）
+- v1.x — 部首/字形分解维度（cjkvi-ids 整合，做表意建模辅助数据）
+- v2.0 — 古文阅读理解 Q&A（GPT 辅助生成）
+
+## 引用
+
+```bibtex
+@misc{chinese-classical-corpus,
+  title  = {Chinese Classical Corpus},
+  author = {dzxr},
+  year   = {2026},
+  url    = {https://huggingface.co/datasets/dzxr/chinese-classical-corpus}
+}
+```
+
+代码仓库：https://github.com/zi6me/chinese-classical-corpus
+
+## License
+
+**CC0 1.0 Universal** — 公有领域，无附加限制。源典籍本就是公有领域；本数据集对它们的结构化处理也以 CC0 释出。
+
+底层依赖各自的 license 详见 [LICENSE](LICENSE)。

corpus.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c35857810bdbf72d964d5e20630d3041d22fe204f501cec9d769b9568506cf7f
+size 55119321

punctuate.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c89eda602321086e73e3fa4f2280c8166d871f580142dea09899c6747fa7fa9
+size 61420969

translate.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f37bf00ab9967ee61ec9daf8a38418374fdab4e016bc322cf471f55b3b460448
+size 707778938