Hypergraph is more powerful. Transform unstructured text into structured knowledge with LLMs. Graphs, hypergraphs, and spatio-temporal extractions — with one command.
Add initial template files for various domains in English and Chinese
yifanfeng97502d24abf081
diff --git a/README.md b/README.mdindex 6fea8e1..3f0f554 100644--- a/README.md+++ b/README.md@@ -1,244 +1,121 @@-# Hyper-Extract 🧠✨+# Hyper-Extract: From Text to Actionable Knowledge Structures 🚀-> **An Intelligent, LLM-Powered Knowledge Extraction Framework**->-> Transform unstructured text into structured Auto containers with the power of Large Language Models.+[](LICENSE)+[](pyproject.toml)------## 🎯 Vision+**Hyper-Extract** is a next-generation knowledge extraction framework designed to transform unstructured text into **searchable, queryable, and actionable** structured knowledge.-**Hyper-Extract** provides a powerful framework for intelligent knowledge extraction. Rather than treating data structures as passive containers, we empower them with **LLM-driven intelligence** to actively extract, merge, and index knowledge from unstructured text.+Unlike traditional extraction tools that only output static data, Hyper-Extract automatically converts extracted structures (Lists, Graphs, Hypergraphs) into an interactive knowledge base. You can directly chat with the structure, perform precise semantic searches, and realize "Extraction as an Application."-Every Auto container in Hyper-Extract follows a unified lifecycle:--1. **Extract** - Intelligent extraction from text with automatic chunking and merging-2. **Merge** - Automatic aggregation with configurable merge strategies-3. **Index** - Semantic vector search capabilities-4. **Serialize** - Persist and reload knowledge with full fidelity+[中文版自述文件 (README_ZH.md)](README_ZH.md)----## ✨ Key Features--### 🤖 LLM-Native Design-- **Deep LangChain Integration**: Every container is initialized with LLM and embedder instances-- **Structured Output**: Leverage `with_structured_output` for type-safe Pydantic schema extraction-- **Intelligent Merging**: First-priority merge strategy for chunk-based extraction+## 🌟 Key Highlights-### 🏗️ Unified Architecture-- **Abstract Base Class**: `BaseAutoType[T]` provides consistent interface across all knowledge types-- **Generic Type Support**: Full Python typing with `Generic[T]` for schema flexibility-- **Lifecycle Management**: Standardized `extract` → `merge` → `search` → `dump/load` pipeline--### 🔍 Semantic Search-- **FAISS Integration**: Fast vector similarity search for knowledge retrieval-- **Automatic Indexing**: Build and maintain indices with lazy loading and dirty tracking-- **Field-Level Search**: Search across all schema fields with metadata preservation--### 📦 Multiple Knowledge Types (Roadmap)-- ✅ **AutoModel**: Universal schema-agnostic knowledge extraction-- ✅ **AutoList**: Collection-based extraction for lists of items-- ✅ **AutoSet**: Unique collection with automatic deduplication-- 🚧 **AutoGraph**: Knowledge graph construction (nodes + edges)-- 🚧 **AutoHypergraph**: Hypergraph representation for complex multi-entity relations-- 🚧 **ExtractionExperience**: Meta-learning from extraction patterns+- **📂 Universal Structure Support**:+ - **Foundational**: List, Set, General Knowledge Graph.+ - **Dynamics**: Temporal Graph (time-series events), Spatial Graph (geographical topology), Spatio-Temporal Graph.+ - **Complex Relations**: Hypergraph (N-ary relations), perfect for modeling group interactions and holistic context.+- **💬 Actionable Knowledge (RAG 2.0)**:+ - Extraction is just the beginning. Hyper-Extract automatically builds vector indices for every extracted entity and relationship.+ - **Chat with Structure**: Query your graph or list directly to get answers backed by structured logic.+- **🛠️ Production-Ready SOTA Algorithms**:+ - Built-in implementations of cutting-edge extraction and RAG paradigms: `Auto-Gen`, `Hyper-RAG`, `Light-RAG`, `GraphRAG`, `CogRAG`, and more.+- **🎭 Massive Domain Templates**:+ - 20+ out-of-the-box templates covering: Finance, Medicine, Law, Gaming, Agriculture, Cybersecurity, Physics, Literature, etc.----## 🛠️ Installation--### Prerequisites-- Python 3.10+-- OpenAI API key (or compatible LLM provider)--### Install with uv (Recommended)-```bash-# Clone the repository-git clone https://github.com/your-username/hyper-extract.git-cd hyper-extract+## 🚀 Quick Start-# Install dependencies with uv-uv sync-```+### Installation-### Install with pip```bash-pip install -r requirements.txt+pip install hyperextract```------## 🚀 Quick Start+### Build an "Intelligent Knowledge Graph" in 3 Steps-### 1. Define Your Knowledge Schema+Hyper-Extract makes complex extraction and interaction dead simple:```pythonfrom pydantic import BaseModel, Field-from typing import List--class ArticleSchema(BaseModel):- """Schema for article knowledge extraction"""- title: str = Field(default="", description="Article title")- author: str = Field(default="", description="Author name")- summary: str = Field(default="", description="Article summary")- key_points: List[str] = Field(default_factory=list, description="Key points")- topics: List[str] = Field(default_factory=list, description="Topic tags")-```--### 2. Initialize Auto Container--```python+from hyperextract.graphs import AutoGraphfrom langchain_openai import ChatOpenAI, OpenAIEmbeddings-from hyperextract import AutoModel--# Initialize LLM and embedder-llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)-embedder = OpenAIEmbeddings(model="text-embedding-3-small")--# Create Auto container-a_article = AutoModel(- data_schema=ArticleSchema,- llm_client=llm,- embedder=embedder,- chunk_size=2000, # Auto-split long texts- max_workers=5, # Concurrent processing- verbose=True-)-```--### 3. Extract Knowledge from Text--```python-article_text = """-Title: The Future of AI-Author: Dr. Jane Smith--Artificial intelligence is transforming industries...-[Your long article text here]-"""--# Extract with automatic chunking and merging-result = a_article.extract(article_text)-print(f"Extraction completed: {result}")--# Access extracted data-print(f"Title: {a_article.data.title}")-print(f"Author: {a_article.data.author}")-print(f"Key points: {a_article.data.key_points}")-```--### 4. Semantic Search--```python-# Build vector index-a_article.build_index()-# Search with natural language-results = a_article.search("What are the main topics?", top_k=3)-for result in results:- print(result)-```--### 5. Save and Load+# 1. Define your world view (Schema)+class entity(BaseModel):+ name: str = Field(description="Name of the entity")+ type: str = Field(description="Category, e.g., Person, Company, Location")++class edge(BaseModel):+ source: str = Field(description="Source entity name")+ target: str = Field(description="Target entity name")+ relation: str = Field(description="The relationship between them")++# 2. Initialize and Build from Text+llm = ChatOpenAI(model="gpt-4o")+embedder = OpenAIEmbeddings()++graph_kb = AutoGraph(+ node_schema=entity,+ edge_schema=edge,+ # Unique key extractors for deduplication and linking+ node_key_extractor=lambda x: x.name,+ edge_key_extractor=lambda x: f"{x.source}_{x.relation}_{x.target}",+ nodes_in_edge_extractor=lambda x: (x.source, x.target),+ llm_client=llm,+ embedder=embedder+)-```python-# Save to folder (includes data + FAISS index)-a_article.dump("./output/my_article")+text = "Apple Inc. was founded by Steve Jobs in California, and it is the maker of the iPhone."+graph_kb.feed_text(text) # Extract knowledge+graph_kb.build_index() # Build vector index for Q&A-# Load from folder-a_article.load("./output/my_article")+# 3. Chat with the Structure (Actionable!)+answer = graph_kb.chat("What is the relationship between Jobs and Apple?")+print(answer.content)```----## 🏛️ Architecture+## 🎨 Domain Templates-### Design Principles+The `hyperextract/templates` directory provides a wealth of domain-specific definitions that you can import directly without writing prompts from scratch:-1. **Template Method Pattern**: `BaseAutoType` defines the extraction skeleton, subclasses customize behavior-2. **First-Priority Merge**: `model_copy(update=model_dump(exclude_none=True))` for conflict resolution-3. **Lazy Indexing**: Build FAISS indices only when needed, track with `_index_dirty` flag-4. **Type Safety**: Leverage Pydantic for runtime validation and IDE support+| Category | Templates | Description |+| :--- | :--- | :--- |+| **General** | `General`, `IText2KG` | Versatile extraction for most daily knowledge. |+| **Science** | `Physics`, `Chemistry`, `Biology` | Capture experiments, elements, and reactions. |+| **Industry** | `Finance`, `Legal`, `Cybersecurity` | Extract compliance, risks, and entity webs. |+| **Entertainment** | `Game`, `Movie`, `Music` | Perfect for story plots and character networks. |+| ... | | 20+ more domains in the [templates folder](hyperextract/templates) |----## 📚 Examples+## 🏗️ Core Components-Check out the `examples/` directory for complete working examples:--- [`auto_model_demo.py`](examples/auto_model_demo.py) - Full AutoModel walkthrough-- More examples coming soon!+- **`hyperextract.core`**: Core abstraction layer handling long-text chunking, parallel processing, and knowledge aggregation.+- **`hyperextract.graphs`**: Various graph components including spatial, temporal, and standard graphs.+- **`hyperextract.hypergraphs`**: Powerful Hypergraph-based RAG implementations for complex multi-entity logic.+- **`hyperextract.utils`**: Essential utilities for logging and merging strategies.----## 🗺️ Roadmap--### Phase 1: Foundation (Current) ✅-- [x] Abstract `BaseAutoType` class-- [x] `AutoModel` implementation-- [x] LangChain integration with structured output-- [x] FAISS vector search-- [x] Serialization (dump/load)--### Phase 2: Specialized Knowledge Types 🚧-- [ ] `AutoGraph` - Full knowledge graph construction-- [ ] `AutoHypergraph` - Hypergraph for complex relations-- [ ] `AutoTree` - Hierarchical tree structures+## 📚 Implemented Algorithms & Paper Demos-### Phase 3: Evolution & Learning 🔮-- [ ] `evolve()` implementation with reasoning-- [ ] Extraction experience learning-- [ ] Confidence scoring and quality metrics-- [ ] Active learning for schema refinement--### Phase 4: Advanced Features 🚀-- [ ] Multi-modal extraction (images, tables, PDFs)-- [ ] Incremental updates and versioning-- [ ] Distributed processing for massive datasets-- [ ] Integration with graph databases (Neo4j, NetworkX)+- **Auto-Graph**: Schema-adaptive graph extraction.+- **Hyper-RAG**: Retrieval Augmented Generation via Hypergraphs.+- **Light-RAG**: Lightweight and efficient graph retrieval.+- **IText2KG / IText2KG-Star**: Iterative Knowledge Graph generation.+- **CogRAG**: Cognition-driven hierarchical retrieval.---## 🤝 Contributing-We welcome contributions! Whether it's:--- 🐛 Bug reports and fixes-- 💡 Feature suggestions-- 📖 Documentation improvements-- 🧪 New knowledge type implementations--Please open an issue or submit a PR. Check out our [contributing guidelines](CONTRIBUTING.md) (coming soon).-------## 📄 License--This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.-------## 🙏 Acknowledgments--- Built with [LangChain](https://github.com/langchain-ai/langchain) for LLM orchestration-- Powered by [Pydantic](https://github.com/pydantic/pydantic) for data validation-- Vector search by [FAISS](https://github.com/facebookresearch/faiss)-- Inspired by the vision of self-evolving knowledge systems-------## 📬 Contact--- **GitHub Issues**: [Create an issue](https://github.com/yifanfeng97/hyper-extract/issues)-- **Email**: [email protected]-------<div align="center">--**⭐ Star this repo if you find it useful! ⭐**--*Building the future of intelligent knowledge extraction, one container at a time.*+We welcome Issues and Pull Requests!-</div>+- **License**: [MIT LICENSE](LICENSE)+- **Support**: If you find this project helpful, please give us a ⭐️!diff --git a/README_ZH.md b/README_ZH.mdnew file mode 100644index 0000000..77a4432--- /dev/null+++ b/README_ZH.md@@ -0,0 +1,121 @@+# Hyper-Extract: 从文本到可交互的知识结构 🚀++[](LICENSE)+[](pyproject.toml)++**Hyper-Extract** 是一个下一代知识提取框架,旨在将非结构化文本转化为**可搜索、可查询、可问答**的结构化知识。++区别于传统的只负责“提取”的工具,Hyper-Extract 将提取出的结构(如列表、图谱、超图)自动转化为可交互的知识库。你可以直接与提取出的结构进行对话,进行精确的语义搜索,实现“提取即应用”。++[English README](README.md)++---++## 🌟 核心特性++- **📂 全模态结构支持**:+ - **基础结构**: List (列表), Set (集合), Graph (通用图谱)。+ - **时空动力学**: Temporal Graph (时序图), Spatial Graph (地理图), Spatio-Temporal Graph (时空图)。+ - **复杂关联**: Hypergraph (超图/多元图),支持多对多复杂关系建模。+- **💬 结构化知识问答 (RAG 2.0)**:+ - 提取不仅仅是终点。Hyper-Extract 自动为提取出的实体和关系建立向量索引。+ - **Chat with Structure**: 直接针对图谱或集合进行问答,获得具有结构化逻辑支撑的答案。+- **🛠️ 工业级算法复现**:+ - 内置了多种 SOTA 提取与检索算法的复现,包括 `Auto-Gen`, `Hyper-RAG`, `Light-RAG`, `GraphRAG`, `CogRAG` 等。+- **🎭 领域专家模版库**:+ - 提供了 20+ 个领域的开箱即用模版,涵盖:金融、医疗、法律、游戏、农业、网络安全、物理、文学等。++---++## 🚀 快速上手++### 安装++```bash+pip install hyperextract+```++### 3步实现“对话知识图谱”++Hyper-Extract 让复杂结构的提取和使用变得异常简单:++```python+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph+from langchain_openai import ChatOpenAI, OpenAIEmbeddings++# 1. 定义你想要提取的结构 (Schema)+class entity(BaseModel):+ name: str = Field(description="实体名称")+ type: str = Field(description="类别,如:人、公司、地点")++class edge(BaseModel):+ source: str = Field(description="源实体")+ target: str = Field(description="目标实体")+ relation: str = Field(description="它们之间的关系")++# 2. 初始化提取器并从文本构建图谱+llm = ChatOpenAI(model="gpt-4o")+embedder = OpenAIEmbeddings()++graph_kb = AutoGraph(+ node_schema=entity,+ edge_schema=edge,+ # 唯一键提取规则,用于去重和关联+ node_key_extractor=lambda x: x.name,+ edge_key_extractor=lambda x: f"{x.source}_{x.relation}_{x.target}",+ nodes_in_edge_extractor=lambda x: (x.source, x.target),+ llm_client=llm,+ embedder=embedder+)++text = "苹果公司由史蒂夫·乔布斯在加利福尼亚州创立,它是著名的iPhone制造商。"+graph_kb.feed_text(text) # 提取知识+graph_kb.build_index() # 构建向量索引 (用于后续问答)++# 3. 针对提取出的结构进行问答 (Actionable!)+answer = graph_kb.chat("乔布斯和苹果公司是什么关系?")+print(answer.content)+```++---++## 🎨 领域模版 (Templates)++本项目在 `hyperextract/templates` 下提供了丰富的领域定义,你可以直接导入使用,无需从头编写 Prompt 或 Schema:++| 领域 | 包含模版 | 描述 |+| :--- | :--- | :--- |+| **通用** | `General`, `IText2KG` | 适用于大多数日常知识提取。 |+| **科学** | `Physics`, `Chemistry`, `Biology` | 捕捉科学实验、元素、反应过程。 |+| **行业** | `Finance`, `Legal`, `Cybersecurity` | 提取合规、风险点、实体关联。 |+| **娱乐** | `Game`, `Movie`, `Music` | 完美支持游戏剧情、角色关系网。 |+| ... | | 更多 20+ 领域见 [templates 目录](hyperextract/templates) |++---++## 🏗️ 核心组件++- **`hyperextract.core`**: 核心抽象层,管理长文本切片、并行处理与知识聚合。+- **`hyperextract.graphs`**: 各类图谱组件(包括时空、地理、标准图谱)。+- **`hyperextract.hypergraphs`**: 强大的超图 RAG 实现,处理超出一对一关系的复杂逻辑。+- **`hyperextract.utils`**: 日志、合并策略等实用工具。++---++## 📚 已复现论文/算法++- **Auto-Graph**: 自动 Schema 适应的图提取。+- **Hyper-RAG**: 基于超图的检索增强生成。+- **Light-RAG**: 轻量化、高效率的图检索。+- **IText2KG / IText2KG-Star**: 迭代式知识图谱生成。+- **CogRAG**: 认知驱动的层次化检索。++---++## 🤝 贡献与反馈++欢迎提交 Issue 或 Pull Request 来完善这个项目!++- **开源协议**: [MIT LICENSE](LICENSE)+- **Star我们**: 如果这个项目对你有帮助,请给一个 ⭐️ 以示支持!diff --git a/hyperextract/templates/__init__.py b/hyperextract/templates/__init__.pyindex e69de29..7f2a477 100644--- a/hyperextract/templates/__init__.py+++ b/hyperextract/templates/__init__.py@@ -0,0 +1,6 @@+from . import en, zh++__all__ = [+ "en",+ "zh",+]diff --git a/hyperextract/templates/en/README.md b/hyperextract/templates/en/README.mdnew file mode 100644index 0000000..c2aa374--- /dev/null+++ b/hyperextract/templates/en/README.md@@ -0,0 +1,50 @@+# Hyper-Extract Domain Templates (English)++Welcome to the **Hyper-Extract Domain Templates** library. This directory contains a robust, flat collection of pre-defined schemas, prompts, and extraction models tailored for specific industries and expertise areas.++## Design Philosophy+- **Flat Discovery**: No deep nesting. Find your domain directly in the top-level folders.+- **Intuitive Naming**: Folders are named after common industry terminology (e.g., `finance`, `medicine`, `game`).+- **Schema-Driven**: Each template provides a professional Pydantic schema to ensure high-quality extraction.++## Domain Index++| Folder | Description | Potential Use Cases |+| :--- | :--- | :--- |+| **`agriculture`** | Farming, livestock, and crop management. | Crop rotation, irrigation systems. |+| **`art`** | Fine arts, museum collections, and provenance. | Artwork catalogs, historical art movements. |+| **`automotive`** | Vehicles, manufacturing, and maintenance. | Car specs, maintenance logs. |+| **`biology`** | Life sciences, genetics, and taxonomy. | Protein interactions, species classification. |+| **`business`** | General corporate profiles and strategy. | Company bios, strategic goals. |+| **`chemistry`** | Molecular structures and reactions. | Chemical compounds, lab experiments. |+| **`construction`** | Building, BIM, and project planning. | Construction milestones, blueprint entities. |+| **`cybersecurity`** | Threat intelligence and network defense. | STIX models, attack vectors. |+| **`education`** | Knowledge points and curriculum. | Course prerequisites, learning paths. |+| **`environment`** | Climate, ecology, and sustainability. | Weather patterns, carbon footprint tracking. |+| **`fashion`** | Garments, trends, and collections. | Design lookbooks, seasonal trends. |+| **`finance`** | Banking, stocks, and investments. | Earnings reports, stock analysis. |+| **`food`** | Culinary, recipes, and nutrition. | Menu planning, recipe breakdown. |+| **`game`** | Game lore, mechanics, and quests. | NPC relations, quest trees, item crafting. |+| **`general`** | Basic encyclopedic knowledge. | General Person/Org/Event extraction. |+| **`history`** | Events, timelines, and figures. | War campaigns, historical biographies. |+| **`legal`** | Law, contracts, and compliance. | Contract obligations, court rulings. |+| **`literature`** | Fiction, poetry, and character arcs. | Character maps, literary themes. |+| **`marketing`** | Campaigns, SEO, and personas. | Customer journeys, brand archetypes. |+| **`medicine`** | Western and clinical medicine. | Diagnosis, symptom mapping, pharma. |+| **`movie`** | Film, TV, and scripts. | Screenplay beats, cast & crew networks. |+| **`music`** | Theory, albums, and discography. | Genre evolution, tracklists. |+| **`news`** | Media and trending topics. | Investigative boards, causality analysis. |+| **`physics`** | Particle physics, astronomy, and circuits. | Celestial systems, circuit paths. |+| **`politics`** | Governance, policy, and geopolitics. | Legislative tracking, election maps. |+| **`real_estate`** | Property, leasing, and zoning. | Property listings, floor plans. |+| **`recruitment`** | HR, resumes, and hiring. | Skill mapping, interview assessment. |+| **`software`** | Architecture, APIs, and DevOps. | Microservices, software dependencies. |+| **`sports`** | Tournaments, matches, and stats. | Match event logs, player profiles. |+| **`supply_chain`**| Logistics and inventory flow. | Supplier tiering, logistics routes. |+| **`travel`** | Tourism and itineraries. | Trip planning, attraction reviews. |++## Usage+Simply import the template you need:+```python+from hyperextract.templates.en.finance import StockGraph+```diff --git a/hyperextract/templates/en/__init__.py b/hyperextract/templates/en/__init__.pyindex e69de29..0050835 100644--- a/hyperextract/templates/en/__init__.py+++ b/hyperextract/templates/en/__init__.py@@ -0,0 +1,9 @@+from .general.knowledge_graph import KnowledgeGraph+from .general.concept_map import ConceptMap+from .general.logic_graph import LogicGraph++__all__ = [+ "KnowledgeGraph",+ "ConceptMap",+ "LogicGraph",+]diff --git a/hyperextract/templates/en/graphs/__init__.py b/hyperextract/templates/en/agriculture/__init__.pysimilarity index 100%rename from hyperextract/templates/en/graphs/__init__.pyrename to hyperextract/templates/en/agriculture/__init__.pydiff --git a/hyperextract/templates/en/hypergraph/__init__.py b/hyperextract/templates/en/art/__init__.pysimilarity index 100%rename from hyperextract/templates/en/hypergraph/__init__.pyrename to hyperextract/templates/en/art/__init__.pydiff --git a/hyperextract/templates/en/sets/__init__.py b/hyperextract/templates/en/automotive/__init__.pysimilarity index 100%rename from hyperextract/templates/en/sets/__init__.pyrename to hyperextract/templates/en/automotive/__init__.pydiff --git a/hyperextract/templates/en/trees/__init__.py b/hyperextract/templates/en/biology/__init__.pysimilarity index 100%rename from hyperextract/templates/en/trees/__init__.pyrename to hyperextract/templates/en/biology/__init__.pydiff --git a/hyperextract/templates/zh/graphs/__init__.py b/hyperextract/templates/en/business/__init__.pysimilarity index 100%rename from hyperextract/templates/zh/graphs/__init__.pyrename to hyperextract/templates/en/business/__init__.pydiff --git a/hyperextract/templates/zh/hypergraph/__init__.py b/hyperextract/templates/en/chemistry/__init__.pysimilarity index 100%rename from hyperextract/templates/zh/hypergraph/__init__.pyrename to hyperextract/templates/en/chemistry/__init__.pydiff --git a/hyperextract/templates/zh/sets/__init__.py b/hyperextract/templates/en/construction/__init__.pysimilarity index 100%rename from hyperextract/templates/zh/sets/__init__.pyrename to hyperextract/templates/en/construction/__init__.pydiff --git a/hyperextract/templates/zh/trees/__init__.py b/hyperextract/templates/en/cybersecurity/__init__.pysimilarity index 100%rename from hyperextract/templates/zh/trees/__init__.pyrename to hyperextract/templates/en/cybersecurity/__init__.pydiff --git a/hyperextract/trees/base.py b/hyperextract/templates/en/education/__init__.pysimilarity index 100%rename from hyperextract/trees/base.pyrename to hyperextract/templates/en/education/__init__.pydiff --git a/hyperextract/templates/en/environment/__init__.py b/hyperextract/templates/en/environment/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/fashion/__init__.py b/hyperextract/templates/en/fashion/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/finance/__init__.py b/hyperextract/templates/en/finance/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/food/__init__.py b/hyperextract/templates/en/food/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/game/__init__.py b/hyperextract/templates/en/game/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/general/__init__.py b/hyperextract/templates/en/general/__init__.pynew file mode 100644index 0000000..aa993f8--- /dev/null+++ b/hyperextract/templates/en/general/__init__.py@@ -0,0 +1,23 @@+"""General purpose knowledge graph templates.++This module provides foundational templates for general information extraction:+- KnowledgeGraph: Factual entity-relationship mapping for news and biographies.+- ConceptMap: Semantic definitions and taxonomies for technical/educational content.+- LogicGraph: Analysis of reasoning, arguments, and causal chains in analytical text.+"""++from .knowledge_graph import KnowledgeGraph, Entity, Relation+from .concept_map import ConceptMap, Concept, ConceptRelation+from .logic_graph import LogicGraph, LogicNode, LogicRelation++__all__ = [+ "KnowledgeGraph",+ "Entity",+ "Relation",+ "ConceptMap",+ "Concept",+ "ConceptRelation",+ "LogicGraph",+ "LogicNode",+ "LogicRelation",+]diff --git a/hyperextract/templates/en/general/concept_map.py b/hyperextract/templates/en/general/concept_map.pynew file mode 100644index 0000000..690d32e--- /dev/null+++ b/hyperextract/templates/en/general/concept_map.py@@ -0,0 +1,69 @@+from typing import List, Optional+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph++# ==============================================================================+# 1. Schema Definitions+# ==============================================================================++class Concept(BaseModel):+ """A conceptual node representing a term, idea, or theory."""+ term: str = Field(description="The technical term or concept name.")+ definition: str = Field(description="A formal definition or explanation of the concept.")+ examples: Optional[List[str]] = Field(default_factory=list, description="Examples that illustrate this concept.")+ attributes: Optional[List[str]] = Field(default_factory=list, description="Key properties or attributes of the concept.")++class ConceptRelation(BaseModel):+ """A semantic relationship between two concepts."""+ source: str = Field(description="The source concept.")+ target: str = Field(description="The target concept.")+ relation_type: str = Field(+ description="Type of relationship (e.g., is-a, part-of, related-to, used-for, instance-of)."+ )++# ==============================================================================+# 2. Prompts+# ==============================================================================++CONCEPT_MAP_PROMPT = (+ "You are an expert in semantic modeling. Extract a concept map from the text, focusing on "+ "definitions, terminologies, and their hierarchical or associative relationships.\n\n"+ "Guidelines:\n"+ "- Identify core terms and provide their precise definitions based on the text.\n"+ "- Use standard semantic relations like 'is-a' (inheritance) or 'part-of' (composition).\n"+ "- Capture illustrative examples if mentioned."+)++CONCEPT_MAP_NODE_PROMPT = (+ "Extract all foundational concepts and technical terms from the text. "+ "For each concept, provide its term, a precise definition, a list of key attributes, "+ "and any specific examples mentioned in the source."+)++CONCEPT_MAP_EDGE_PROMPT = (+ "Establish semantic relationships between the provided concepts. "+ "Focus on hierarchical taxonomies (is-a, part-of) and functional associations (used-for, instance-of). "+ "Ensure edges logically connect the defined concepts."+)++# ==============================================================================+# 3. Template Class+# ==============================================================================++class ConceptMap(AutoGraph[Concept, ConceptRelation]):+ """+ A template for building conceptual graphs and taxonomies.+ Ideal for technical documentation, educational materials, and glossaries.+ """+ def __init__(self, **kwargs):+ super().__init__(+ node_schema=Concept,+ edge_schema=ConceptRelation,+ node_key_extractor=lambda x: x.term.strip(),+ edge_key_extractor=lambda x: f"{x.source.strip()}--({x.relation_type.lower()})-->{x.target.strip()}",+ nodes_in_edge_extractor=lambda x: (x.source.strip(), x.target.strip()),+ prompt=CONCEPT_MAP_PROMPT,+ prompt_for_node_extraction=CONCEPT_MAP_NODE_PROMPT,+ prompt_for_edge_extraction=CONCEPT_MAP_EDGE_PROMPT,+ **kwargs+ )diff --git a/hyperextract/templates/en/general/knowledge_graph.py b/hyperextract/templates/en/general/knowledge_graph.pynew file mode 100644index 0000000..3d3a3bd--- /dev/null+++ b/hyperextract/templates/en/general/knowledge_graph.py@@ -0,0 +1,71 @@+from typing import List, Optional+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph++# ==============================================================================+# 1. Schema Definitions+# ==============================================================================++class Entity(BaseModel):+ """A general entity representing a person, organization, location, or object."""+ name: str = Field(description="The primary name of the entity.")+ category: str = Field(+ description="Category of the entity (e.g., Person, Organization, Location, Event, Object)."+ )+ description: Optional[str] = Field(+ description="A concise summary of the entity and its role in the text."+ )++class Relation(BaseModel):+ """A factual relationship between two entities."""+ source: str = Field(description="The name of the source entity.")+ target: str = Field(description="The name of the target entity.")+ relation: str = Field(description="The type of relationship (e.g., Works for, Located in, Born in, Acquired).")+ details: Optional[str] = Field(description="Additional context or details about this relationship.")++# ==============================================================================+# 2. Prompts+# ==============================================================================++KNOWLEDGE_GRAPH_PROMPT = (+ "You are an expert knowledge extraction system. Your task is to extract a factual knowledge graph "+ "from the provided text. Focus on identifying key entities (People, Organizations, Locations, and relevant Objects) "+ "and the explicit relationships between them.\n\n"+ "Guidelines:\n"+ "- Extract distinct entitles and their factual properties.\n"+ "- Map relationships that describe how entities interact or are connected.\n"+ "- Use clear, concise language for descriptions and relations."+)++KNOWLEDGE_GRAPH_NODE_PROMPT = (+ "Extract all key entities from the text. Focus on identifying People, Organizations, Locations, and important Objects. "+ "Provide a name, a category, and a concise description for each entity."+)++KNOWLEDGE_GRAPH_EDGE_PROMPT = (+ "Identify factual relationships between the following listed entities based on the text. "+ "Focus on interactions like 'works for', 'located in', 'acquired', or 'partnered with'. "+ "Do not hallucinate entities not in the provided list."+)++# ==============================================================================+# 3. Template Class+# ==============================================================================++class KnowledgeGraph(AutoGraph[Entity, Relation]):+ """+ A foundational knowledge graph template for factual extraction.+ Best for news, biographies, and descriptive encyclopedic content.+ """+ def __init__(self, **kwargs):+ super().__init__(+ node_schema=Entity,+ edge_schema=Relation,+ node_key_extractor=lambda x: x.name.strip(),+ edge_key_extractor=lambda x: f"{x.source.strip()}--[{x.relation.lower()}]-->{x.target.strip()}",+ nodes_in_edge_extractor=lambda x: (x.source.strip(), x.target.strip()),+ prompt=KNOWLEDGE_GRAPH_PROMPT,+ prompt_for_node_extraction=KNOWLEDGE_GRAPH_NODE_PROMPT,+ prompt_for_edge_extraction=KNOWLEDGE_GRAPH_EDGE_PROMPT,+ **kwargs+ )diff --git a/hyperextract/templates/en/general/logic_graph.py b/hyperextract/templates/en/general/logic_graph.pynew file mode 100644index 0000000..56362ab--- /dev/null+++ b/hyperextract/templates/en/general/logic_graph.py@@ -0,0 +1,67 @@+from typing import List, Optional+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph++# ==============================================================================+# 1. Schema Definitions+# ==============================================================================++class LogicNode(BaseModel):+ """A node representing a claim, evidence, or premise in an argument."""+ statement: str = Field(description="The central claim, fact, or observation.")+ node_type: str = Field(+ description="Type of node: Claim (main point), Evidence (supporting data), Premise (underlying assumption)."+ )+ source_attribution: Optional[str] = Field(description="Where or who this logic point originates from.")++class LogicRelation(BaseModel):+ """The logical connection between two statements."""+ source: str = Field(description="The source statement.")+ target: str = Field(description="The target statement.")+ inference: str = Field(+ description="The logical link: Supports, Contradicts, Proves, Leads to, Explains."+ )++# ==============================================================================+# 2. Prompts+# ==============================================================================++LOGIC_GRAPH_PROMPT = (+ "You are a logical analyst. Extract the reasoning structure from the text into a directed graph.\n\n"+ "Guidelines:\n"+ "- Identify the main claims (conclusions) and the evidence or premises that lead to them.\n"+ "- Explicitly map how one statement supports or contradicts another.\n"+ "- Do not just extract facts; extract the 'argument' and 'reasoning' flow."+)++LOGIC_GRAPH_NODE_PROMPT = (+ "Extract the individual building blocks of the argument. Identify core claims, supporting evidence, "+ "and underlying premises. For each, determine its type and note any source attribution mentioned."+)++LOGIC_GRAPH_EDGE_PROMPT = (+ "Map the logical flow between the extracted claims and evidence. Focus on how statements "+ "Support, Contradict, or Lead to one another. Ensure every edge represents a step in the reasoning chain."+)++# ==============================================================================+# 3. Template Class+# ==============================================================================++class LogicGraph(AutoGraph[LogicNode, LogicRelation]):+ """+ A template for analyzing reasoning, arguments, and causal chains.+ Ideal for analytical reports, debate transcripts, and scientific reasoning.+ """+ def __init__(self, **kwargs):+ super().__init__(+ node_schema=LogicNode,+ edge_schema=LogicRelation,+ node_key_extractor=lambda x: x.statement.strip(),+ edge_key_extractor=lambda x: f"{x.source.strip()}--({x.inference.lower()})-->{x.target.strip()}",+ nodes_in_edge_extractor=lambda x: (x.source.strip(), x.target.strip()),+ prompt=LOGIC_GRAPH_PROMPT,+ prompt_for_node_extraction=LOGIC_GRAPH_NODE_PROMPT,+ prompt_for_edge_extraction=LOGIC_GRAPH_EDGE_PROMPT,+ **kwargs+ )diff --git a/hyperextract/templates/en/history/__init__.py b/hyperextract/templates/en/history/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/legal/__init__.py b/hyperextract/templates/en/legal/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/literature/__init__.py b/hyperextract/templates/en/literature/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/marketing/__init__.py b/hyperextract/templates/en/marketing/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/medicine/__init__.py b/hyperextract/templates/en/medicine/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/movie/__init__.py b/hyperextract/templates/en/movie/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/music/__init__.py b/hyperextract/templates/en/music/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/news/__init__.py b/hyperextract/templates/en/news/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/physics/__init__.py b/hyperextract/templates/en/physics/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/politics/__init__.py b/hyperextract/templates/en/politics/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/real_estate/__init__.py b/hyperextract/templates/en/real_estate/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/recruitment/__init__.py b/hyperextract/templates/en/recruitment/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/software/__init__.py b/hyperextract/templates/en/software/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/sports/__init__.py b/hyperextract/templates/en/sports/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/supply_chain/__init__.py b/hyperextract/templates/en/supply_chain/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/en/travel/__init__.py b/hyperextract/templates/en/travel/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/README.md b/hyperextract/templates/zh/README.mdnew file mode 100644index 0000000..c634969--- /dev/null+++ b/hyperextract/templates/zh/README.md@@ -0,0 +1,53 @@+# Hyper-Extract 领域知识模板 (中文版)++欢迎使用 **Hyper-Extract 领域模板库**。这里提供了一系列针对中文语境和国内行业标准优化的提取模板。++## 设计理念+- **直觉导航**:采用扁平化目录结构,拒绝深层嵌套。+- **行业深度**:每个模板均由特定领域的知识结构(Schema)和提示词(Prompt)驱动。+- **文化融合**:在保持与国际标准接轨的同时,特别加入了如中医(TCM)、武侠(Wuxia)等中国特色文化领域。++## 领域索引++| 目录名称 | 说明 | 应用场景示例 |+| :--- | :--- | :--- |+| **`tcm`** | **中医 (特色)** | 中草药方剂、经络穴位、辨证施治过程等。 |+| **`wuxia`** | **武侠/修真 (特色)** | 门派恩怨、功法境界、修仙等级体系、法宝神兵。 |+| **`general`** | 通用百科 | 人物、机构、事件、地点等基础实体的快速抽取。 |+| **`finance`** | 金融投资 | A股/港股研报、基金招募说明书、上市公司季报。 |+| **`legal`** | 法律合规 | 合同关键要素(义务、权利)、案情调查、法条引用。 |+| **`education`** | 教育学术 | K12知识点图谱、课程大纲依赖关系、学术论文脉络。 |+| **`business`** | 商业通用 | 公司全画像、股权穿透结构、商业战略目标。 |+| **`medicine`** | 医学健康 | 疾病症状映射、临床诊疗路径、西药说明书结构化。 |+| **`food`** | 美食餐饮 | 中餐菜谱(刀工、火候)、食材配比、八大菜系特征。 |+| **`history`** | 历史文化 | 朝代更迭时间线、重大战役、历史人物生平关系。 |+| **`literature`** | 文学 | 小说人物关系网、文学流派演变、古诗词意象关联。 |+| **`software`** | 软件与IT | 系统架构图、接口定义、代码依赖关系、DevOps流程。 |+| **`game`** | 游戏娱乐 | 游戏世界观设定、任务流向树、物品合成与掉落表。 |+| **`sports`** | 体育竞技 | 赛事文字直播提取、球员身价与转会网络、战术阵型。 |+| **`travel`** | 旅游出行 | 游记攻略结构化、景点关系图、旅行行程规划。 |+| **`recruitment`**| 招聘简历 | 简历经历解析、技能标签映射、岗位/人才画像匹配。 |+| **`marketing`** | 市场营销 | 用户旅程地图 (CJM)、品牌定位分析、营销活动链路。 |+| **`supply_chain`**| 物流/供应链 | 供应链上下游追溯、物流节点关系、库存流转图。 |+| **`agriculture`** | 农业与养殖 | 作物生长周期、病虫害防治、畜牧谱系与饲养标准。 |+| **`art`** | 艺术与博物馆 | 艺术品流传有序图、艺术流派、博物馆馆藏关系。 |+| **`automotive`** | 汽车工业 | 汽车参数配置、维保手册提取、零部件组成。 |+| **`biology`** | 生物科学 | 生物分类学、蛋白质相互作用、基因调控网络。 |+| **`chemistry`** | 化学科学 | 分子结构、化学反应方程式路径、材料物化特性。 |+| **`physics`** | 物理与天文 | 天体运行系统、量子物理现象、电路逻辑连接图。 |+| **`construction`**| 建筑工程 | 建筑信息模型 (BIM) 实体、施工进度节点、材料明细。 |+| **`cybersecurity`**| 网络安全 | 攻击路径图、威胁情报 STIX 模型、系统漏洞分析。 |+| **`environment`** | 环境与气候 | 气象变化趋势、碳排放追踪、生态系统食物网。 |+| **`fashion`** | 时尚设计 | 时装系列设计元素、流行趋势周期、单品构成。 |+| **`movie`** | 影视 | 剧本节拍器拆解、演员与幕后人员网络、影评情感。 |+| **`music`** | 音乐 | 乐理结构、专辑曲目分类、流派演进。 |+| **`news`** | 新闻传媒 | 事件因果链、新闻专题聚类、深度报道事实板。 |+| **`politics`** | 政治与政府 | 政策影响分析、地缘政治关系网、选举数据地图。 |+| **`real_estate`** | 房地产 | 房源户型特征、租赁协议要素、土地规划关系。 |++*(注:英文版对应的 31 个通用领域在中文版中均有映射文件夹,中医与武侠为中文版独有)*++## 快速上手+```python+from hyperextract.templates.zh.tcm import HerbGraph+```diff --git a/hyperextract/templates/zh/__init__.py b/hyperextract/templates/zh/__init__.pyindex e69de29..0050835 100644--- a/hyperextract/templates/zh/__init__.py+++ b/hyperextract/templates/zh/__init__.py@@ -0,0 +1,9 @@+from .general.knowledge_graph import KnowledgeGraph+from .general.concept_map import ConceptMap+from .general.logic_graph import LogicGraph++__all__ = [+ "KnowledgeGraph",+ "ConceptMap",+ "LogicGraph",+]diff --git a/hyperextract/templates/zh/agriculture/__init__.py b/hyperextract/templates/zh/agriculture/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/art/__init__.py b/hyperextract/templates/zh/art/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/automotive/__init__.py b/hyperextract/templates/zh/automotive/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/biology/__init__.py b/hyperextract/templates/zh/biology/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/business/__init__.py b/hyperextract/templates/zh/business/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/chemistry/__init__.py b/hyperextract/templates/zh/chemistry/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/construction/__init__.py b/hyperextract/templates/zh/construction/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/cybersecurity/__init__.py b/hyperextract/templates/zh/cybersecurity/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/education/__init__.py b/hyperextract/templates/zh/education/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/environment/__init__.py b/hyperextract/templates/zh/environment/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/fashion/__init__.py b/hyperextract/templates/zh/fashion/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/finance/__init__.py b/hyperextract/templates/zh/finance/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/food/__init__.py b/hyperextract/templates/zh/food/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/game/__init__.py b/hyperextract/templates/zh/game/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/general/__init__.py b/hyperextract/templates/zh/general/__init__.pynew file mode 100644index 0000000..8a8e6be--- /dev/null+++ b/hyperextract/templates/zh/general/__init__.py@@ -0,0 +1,23 @@+"""领域通用知识图谱模板。++本模块提供基础的信息提取模板,适用于多种场景:+- KnowledgeGraph: 事实实体-关系映射,适用于新闻、传记和百科。+- ConceptMap: 语义定义与分类体系,适用于技术文档和教育材料。+- LogicGraph: 推理分析、论证过程及因果链,适用于分析性文本。+"""++from .knowledge_graph import KnowledgeGraph, Entity, Relation+from .concept_map import ConceptMap, Concept, ConceptRelation+from .logic_graph import LogicGraph, LogicNode, LogicRelation++__all__ = [+ "KnowledgeGraph",+ "Entity",+ "Relation",+ "ConceptMap",+ "Concept",+ "ConceptRelation",+ "LogicGraph",+ "LogicNode",+ "LogicRelation",+]diff --git a/hyperextract/templates/zh/general/concept_map.py b/hyperextract/templates/zh/general/concept_map.pynew file mode 100644index 0000000..af793dd--- /dev/null+++ b/hyperextract/templates/zh/general/concept_map.py@@ -0,0 +1,66 @@+from typing import List, Optional+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph++# ==============================================================================+# 1. Schema 定义 (Schema Definitions)+# ==============================================================================++class Concept(BaseModel):+ """一个代表术语、想法或理论的概念节点。"""+ term: str = Field(description="技术术语或概念的名称。")+ definition: str = Field(description="该概念的正式定义或解释。")+ examples: Optional[List[str]] = Field(default_factory=list, description="说明该概念的示例。")+ attributes: Optional[List[str]] = Field(default_factory=list, description="该概念的关键特征或属性。")++class ConceptRelation(BaseModel):+ """两个概念之间的语义关系。"""+ source: str = Field(description="源概念。")+ target: str = Field(description="目标概念。")+ relation_type: str = Field(+ description="关系类型(例如:是一种 (is-a), 属于 (part-of), 相关于, 用于, 为...的实例)。"+ )++# ==============================================================================+# 2. 提示词 (Prompts)+# ==============================================================================++CONCEPT_MAP_PROMPT = (+ "你是一位语义建模专家。请从文本中提取概念图(Concept Map),重点关注术语定义及其层级或关联关系。\n\n"+ "提取指南:\n"+ "- 识别核心术语,并基于文本提供精确定义。\n"+ "- 使用标准的语义关系词,如“是一种”(继承关系)或“属于”(组成关系)。\n"+ "- 如果文本提到了说明性的示例,请一并捕获。"+)++CONCEPT_MAP_NODE_PROMPT = (+ "请从文本中提取所有基础概念和技术术语。对于每个概念,提供其术语名称、精确定义、关键属性列表,"+ "以及文本中提到的任何具体示例。"+)++CONCEPT_MAP_EDGE_PROMPT = (+ "在提供的概念之间建立语义关系。重点关注层级分类(是一种、属于)以及功能性关联(用于、为...的实例)。"+ "确保边逻辑上连接了已定义的各个概念。"+)++# ==============================================================================+# 3. 模板类 (Template Class)+# ==============================================================================++class ConceptMap(AutoGraph[Concept, ConceptRelation]):+ """+ 用于构建概念图谱和分类体系的模板。+ 适用于技术文档、教育材料、术语表解析等场景。+ """+ def __init__(self, **kwargs):+ super().__init__(+ node_schema=Concept,+ edge_schema=ConceptRelation,+ node_key_extractor=lambda x: x.term.strip(),+ edge_key_extractor=lambda x: f"{x.source.strip()}--({x.relation_type.lower()})-->{x.target.strip()}",+ nodes_in_edge_extractor=lambda x: (x.source.strip(), x.target.strip()),+ prompt=CONCEPT_MAP_PROMPT,+ prompt_for_node_extraction=CONCEPT_MAP_NODE_PROMPT,+ prompt_for_edge_extraction=CONCEPT_MAP_EDGE_PROMPT,+ **kwargs+ )diff --git a/hyperextract/templates/zh/general/knowledge_graph.py b/hyperextract/templates/zh/general/knowledge_graph.pynew file mode 100644index 0000000..f853d4c--- /dev/null+++ b/hyperextract/templates/zh/general/knowledge_graph.py@@ -0,0 +1,68 @@+from typing import List, Optional+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph++# ==============================================================================+# 1. Schema 定义 (Schema Definitions)+# ==============================================================================++class Entity(BaseModel):+ """通用实体,代表人物、组织、地点、物体或核心概念。"""+ name: str = Field(description="实体的名称。")+ category: str = Field(+ description="实体的类别(例如:人物、组织、地理位置、概念、事件、物体)。"+ )+ description: Optional[str] = Field(+ description="该实体的简要描述及其在文本中的作用。"+ )++class Relation(BaseModel):+ """两个实体之间的事实关系。"""+ source: str = Field(description="源实体的名。")+ target: str = Field(description="目标实体的名称。")+ relation: str = Field(description="关系类型(例如:任职于、位于、创立、收购)。")+ details: Optional[str] = Field(description="关于该关系的额外上下文或详细描述。")++# ==============================================================================+# 2. 提示词 (Prompts)+# ==============================================================================++KNOWLEDGE_GRAPH_PROMPT = (+ "你是一个专业的知识提取专家。你的任务是从提供的文本中提取一个事实知识图谱。"+ "重点识别关键实体(人物、组织、地理位置、重要物体)以及它们之间明确的关系。\n\n"+ "提取指南:\n"+ "- 准确性:从文本中提取明确的实体及其事实属性。\n"+ "- 关系映射:描述实体之间如何互动或连接。\n"+ "- 简洁性:使用清晰、简洁的语言进行描述和关系定义。"+)++KNOWLEDGE_GRAPH_NODE_PROMPT = (+ "请从文本中提取所有关键实体。重点识别人物、组织、地理位置以及重要的物体。"+ "为每个实体提供名称、类别以及简明扼要的描述。"+)++KNOWLEDGE_GRAPH_EDGE_PROMPT = (+ "基于文本识别以下已知实体之间的事实关系。重点关注如“任职于”、“位于”、“创立”或“收购”等互动。"+ "请勿虚构不在列表中的实体。"+)++# ==============================================================================+# 3. 模板类 (Template Class)+# ==============================================================================++class KnowledgeGraph(AutoGraph[Entity, Relation]):+ """+ 通用事实知识图谱模板,适用于新闻报道、百科条目及传记。+ """+ def __init__(self, **kwargs):+ super().__init__(+ node_schema=Entity,+ edge_schema=Relation,+ node_key_extractor=lambda x: x.name.strip(),+ edge_key_extractor=lambda x: f"{x.source.strip()}--[{x.relation.lower()}]-->{x.target.strip()}",+ nodes_in_edge_extractor=lambda x: (x.source.strip(), x.target.strip()),+ prompt=KNOWLEDGE_GRAPH_PROMPT,+ prompt_for_node_extraction=KNOWLEDGE_GRAPH_NODE_PROMPT,+ prompt_for_edge_extraction=KNOWLEDGE_GRAPH_EDGE_PROMPT,+ **kwargs+ )diff --git a/hyperextract/templates/zh/general/logic_graph.py b/hyperextract/templates/zh/general/logic_graph.pynew file mode 100644index 0000000..85718a7--- /dev/null+++ b/hyperextract/templates/zh/general/logic_graph.py@@ -0,0 +1,65 @@+from typing import List, Optional+from pydantic import BaseModel, Field+from hyperextract.graphs import AutoGraph++# ==============================================================================+# 1. Schema 定义 (Schema Definitions)+# ==============================================================================++class LogicNode(BaseModel):+ """代表论点中的主张、证据或前提的节点。"""+ statement: str = Field(description="核心主张、事实或观察结果。")+ node_type: str = Field(+ description="节点类型:主张 (Claim), 证据 (Evidence), 前提 (Premise)。"+ )+ source_attribution: Optional[str] = Field(description="该逻辑点的来源或出处。")++class LogicRelation(BaseModel):+ """两个陈述之间的逻辑连接。"""+ source: str = Field(description="源陈述。")+ target: str = Field(description="目标陈述。")+ inference: str = Field(+ description="逻辑链路:支持 (Supports), 反驳 (Contradicts), 证明, 导致, 解释。"+ )++# ==============================================================================+# 2. 提示词 (Prompts)+# ==============================================================================++LOGIC_GRAPH_PROMPT = (+ "你是一位逻辑分析师。请将文本中的论证结构提取并转化为一个有向图。\n\n"+ "提取指南:\n"+ "- 识别主要主张(结论)以及引向这些结论的证据或前提。\n"+ "- 明确映射一个陈述是如何支持或反驳另一个陈述的。\n"+ "- 不仅要提取事实,更要提取“论证”和“推理”的流向。"+)++LOGIC_GRAPH_NODE_PROMPT = (+ "请提取论证中的各个基本构成要素。识别核心主张、支持性证据和底层前提。对于每一项,确定其类型并记录提及的任何来源出处。"+)++LOGIC_GRAPH_EDGE_PROMPT = (+ "映射提取出的主张与证据之间的逻辑流向。重点关注各个陈述之间是如何相互 支持、反驳 或 导致 彼此的。确保每条边都代表推理链中的一个步骤。"+)++# ==============================================================================+# 3. 模板类 (Template Class)+# ==============================================================================++class LogicGraph(AutoGraph[LogicNode, LogicRelation]):+ """+ 用于分析论文推理、论证和因果链的模板。+ 适用于分析报告、辩论内容、科研推理及逻辑分析。+ """+ def __init__(self, **kwargs):+ super().__init__(+ node_schema=LogicNode,+ edge_schema=LogicRelation,+ node_key_extractor=lambda x: x.statement.strip(),+ edge_key_extractor=lambda x: f"{x.source.strip()}--({x.inference.lower()})-->{x.target.strip()}",+ nodes_in_edge_extractor=lambda x: (x.source.strip(), x.target.strip()),+ prompt=LOGIC_GRAPH_PROMPT,+ prompt_for_node_extraction=LOGIC_GRAPH_NODE_PROMPT,+ prompt_for_edge_extraction=LOGIC_GRAPH_EDGE_PROMPT,+ **kwargs+ )diff --git a/hyperextract/templates/zh/history/__init__.py b/hyperextract/templates/zh/history/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/legal/__init__.py b/hyperextract/templates/zh/legal/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/literature/__init__.py b/hyperextract/templates/zh/literature/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/marketing/__init__.py b/hyperextract/templates/zh/marketing/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/medicine/__init__.py b/hyperextract/templates/zh/medicine/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/movie/__init__.py b/hyperextract/templates/zh/movie/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/music/__init__.py b/hyperextract/templates/zh/music/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/news/__init__.py b/hyperextract/templates/zh/news/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/physics/__init__.py b/hyperextract/templates/zh/physics/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/politics/__init__.py b/hyperextract/templates/zh/politics/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/real_estate/__init__.py b/hyperextract/templates/zh/real_estate/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/recruitment/__init__.py b/hyperextract/templates/zh/recruitment/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/software/__init__.py b/hyperextract/templates/zh/software/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/sports/__init__.py b/hyperextract/templates/zh/sports/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/supply_chain/__init__.py b/hyperextract/templates/zh/supply_chain/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/tcm/__init__.py b/hyperextract/templates/zh/tcm/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/travel/__init__.py b/hyperextract/templates/zh/travel/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/templates/zh/wuxia/__init__.py b/hyperextract/templates/zh/wuxia/__init__.pynew file mode 100644index 0000000..e69de29diff --git a/hyperextract/trees/__init__.py b/hyperextract/trees/__init__.pydeleted file mode 100644index b0ce327..0000000--- a/hyperextract/trees/__init__.py+++ /dev/null@@ -1,10 +0,0 @@-"""Tree structures for hierarchical data extraction.--This module provides tree-based Auto types for extracting hierarchical-and nested data structures from unstructured text.-"""--# Tree implementations will be added here as they are developed-# from .base import AutoTree--__all__ = []