# yh7dfhvbk4-code/ryy-projects architecture

> Architecture diagram of the GitHub repository yh7dfhvbk4-code/ryy-projects, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-04). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/yh7dfhvbk4-code/ryy-projects
- Repository: https://github.com/yh7dfhvbk4-code/ryy-projects

## Overview

This repository is a collection of five separate projects, not one integrated application. Their principal workflows are: submit ancient text for entity/relation extraction and graph construction; build a personal-document index and query it with RAG; use parking-management web interfaces; explore AI topics through a website; and run CNN-based zebra-crossing/obstacle recognition. The古籍 pipeline and RAG internals are supported by excerpts; other projects are represented from the file tree and README only, so their internal wiring is not inferred.

Main entry points: [`RAG/rag/engine.py`](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/engine.py) (RAG引擎), [`GuJi古籍翻译/src/pipeline.py`](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/pipeline.py) (文本流水线), [`RAG/rag/ingestion/pipeline.py`](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/pipeline.py) (索引构建), [`RAG/rag/retrieval/hybrid.py`](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/hybrid.py) (混合检索)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_guji["古籍语义与图谱"]
  node_guji_api["Web API<br/>[web_api.py]"]
  node_guji_pipeline["文本流水线<br/>[pipeline.py]"]
  node_cleaner["文本清洗<br/>[text_cleaner.py]"]
  node_tokenizer["分词断句<br/>[tokenizer.py]"]
  node_ner_predictor["实体识别<br/>[predictor.py]"]
  node_ner_model["BERT-CRF模型<br/>[model.py]"]
  node_ner_dataset["NER数据集<br/>[dataset.py]"]
  node_few_shot["少样本处理<br/>[few_shot.py]"]
  node_relation_extract["关系抽取"]
  node_knowledge_graph["知识图谱构建<br/>[knowledge_graph.py]"]
end

subgraph group_rag["个人知识库 RAG"]
  node_rag_engine["RAG引擎<br/>[engine.py]"]
  node_ingest["索引构建<br/>[pipeline.py]"]
  node_loaders["文档加载<br/>[loaders.py]"]
  node_chunking["文本分片<br/>[chunking.py]"]
  node_vector_store[("向量存储<br/>[vector_store.py]")]
  node_embeddings["文本嵌入<br/>[embeddings.py]"]
  node_bm25[("BM25索引<br/>[bm25.py]")]
  node_hybrid["混合检索<br/>[hybrid.py]"]
  node_reranker["结果重排<br/>[reranker.py]"]
  node_prompts["上下文与提示<br/>[prompts.py]"]
  node_llm["LLM客户端<br/>[llm.py]"]
end

subgraph group_parking["停车管理"]
  node_parking_frontend["前端应用<br/>[app.js]"]
  node_parking_backend["后端应用<br/>[app.py]"]
end

subgraph group_aiweb["AI介绍网站"]
  node_site_main["网站主脚本<br/>[main.js]"]
  node_site_components["页面组件<br/>[components.js]"]
  node_site_chat["AI聊天界面<br/>[ai-chat.js]"]
end

subgraph group_vision["CNN视觉识别"]
  node_vision_cnn["CNN识别模型<br/>[cnn模型障碍物识别.py]"]
end

node_text_user(("文本用户"))
node_neo4j[("Neo4j图数据库")]
node_rag_user(("知识库用户"))
node_llm_service["兼容LLM服务"]
node_parking_user(("停车用户"))
node_site_user(("网站访客"))
node_vision_user(("识别使用者"))

node_text_user -.->|"提交文本"| node_guji_api
node_guji_api -->|"调用流水线"| node_guji_pipeline
node_guji_pipeline -->|"清洗文本"| node_cleaner
node_guji_pipeline -->|"分词断句"| node_tokenizer
node_guji_pipeline -->|"识别实体"| node_ner_predictor
node_ner_predictor -->|"调用模型"| node_ner_model
node_guji_pipeline -->|"抽取关系"| node_relation_extract
node_guji_pipeline -->|"构建图谱"| node_knowledge_graph
node_knowledge_graph -.->|"写入图数据"| node_neo4j
node_ner_dataset -.->|"提供训练数据"| node_ner_model
node_few_shot -.->|"增强样本"| node_ner_dataset
node_rag_user -.->|"提问建库"| node_rag_engine
node_rag_engine -->|"构建索引"| node_ingest
node_ingest -->|"加载文档"| node_loaders
node_ingest -->|"切分文档"| node_chunking
node_ingest -->|"写入分片"| node_vector_store
node_ingest -->|"重建索引"| node_bm25
node_vector_store -->|"向量编码"| node_embeddings
node_rag_engine -->|"加载索引"| node_bm25
node_rag_engine -->|"混合召回"| node_hybrid
node_hybrid -->|"向量检索"| node_vector_store
node_hybrid -->|"关键词检索"| node_bm25
node_rag_engine -->|"重排候选"| node_reranker
node_rag_engine -->|"组装上下文"| node_prompts
node_rag_engine -->|"生成回答"| node_llm
node_llm -->|"请求补全"| node_llm_service
node_parking_user -.->|"使用界面"| node_parking_frontend
node_parking_frontend -.->|"请求服务"| node_parking_backend
node_site_user -.->|"访问网站"| node_site_main
node_site_user -.->|"使用聊天"| node_site_chat
node_vision_user -.->|"提交图像"| node_vision_cnn

click node_guji_api "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/web_api.py"
click node_guji_pipeline "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/pipeline.py"
click node_cleaner "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/text_cleaner.py"
click node_tokenizer "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/tokenizer.py"
click node_ner_predictor "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/ner/predictor.py"
click node_ner_model "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/ner/model.py"
click node_ner_dataset "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/ner/dataset.py"
click node_few_shot "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/few_shot.py"
click node_relation_extract "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/relation_extractor.py"
click node_knowledge_graph "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/knowledge_graph.py"
click node_rag_engine "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/engine.py"
click node_ingest "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/pipeline.py"
click node_loaders "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/loaders.py"
click node_chunking "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/chunking.py"
click node_vector_store "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/vector_store.py"
click node_embeddings "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/embeddings.py"
click node_bm25 "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/bm25.py"
click node_hybrid "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/hybrid.py"
click node_reranker "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/reranker.py"
click node_prompts "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/generation/prompts.py"
click node_llm "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/generation/llm.py"
click node_parking_frontend "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/%E5%81%9C%E8%BD%A6%E7%AE%A1%E7%90%86%E7%B3%BB%E7%BB%9F/frontend/js/app.js"
click node_parking_backend "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/%E5%81%9C%E8%BD%A6%E7%AE%A1%E7%90%86%E7%B3%BB%E7%BB%9F/backend/app.py"
click node_site_main "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/AI%E6%8A%80%E6%9C%AF%E4%BB%8B%E7%BB%8D%E7%BD%91%E9%A1%B5/js/main.js"
click node_site_components "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/AI%E6%8A%80%E6%9C%AF%E4%BB%8B%E7%BB%8D%E7%BD%91%E9%A1%B5/js/components.js"
click node_site_chat "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/AI%E6%8A%80%E6%9C%AF%E4%BB%8B%E7%BB%8D%E7%BD%91%E9%A1%B5/js/ai-chat.js"
click node_vision_cnn "https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/cnn%20%E6%96%91%E9%A9%AC%E7%BA%BF%E8%AF%86%E5%88%AB/cnn%E6%A8%A1%E5%9E%8B%E9%9A%9C%E7%A2%8D%E7%89%A9%E8%AF%86%E5%88%AB.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_guji_api,node_guji_pipeline,node_cleaner,node_tokenizer,node_ner_predictor,node_ner_model,node_ner_dataset,node_few_shot,node_relation_extract,node_knowledge_graph toneBlue
class node_rag_engine,node_ingest,node_loaders,node_chunking,node_vector_store,node_embeddings,node_bm25,node_hybrid,node_reranker,node_prompts,node_llm,node_neo4j toneAmber
class node_parking_frontend,node_parking_backend toneMint
class node_site_main,node_site_components,node_site_chat toneRose
class node_vision_cnn,node_text_user,node_rag_user,node_llm_service,node_parking_user,node_site_user,node_vision_user toneIndigo
```

## Components

### 古籍语义与图谱

- [Web API](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/web_api.py): `GuJi古籍翻译/src/web_api.py`
- [文本流水线](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/pipeline.py): `GuJi古籍翻译/src/pipeline.py`
- [文本清洗](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/text_cleaner.py): `GuJi古籍翻译/src/text_cleaner.py`
- [分词断句](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/tokenizer.py): `GuJi古籍翻译/src/tokenizer.py`
- [实体识别](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/ner/predictor.py): `GuJi古籍翻译/src/ner/predictor.py`
- [BERT-CRF模型](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/ner/model.py): `GuJi古籍翻译/src/ner/model.py`
- [NER数据集](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/ner/dataset.py): `GuJi古籍翻译/src/ner/dataset.py`
- [少样本处理](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/few_shot.py): `GuJi古籍翻译/src/few_shot.py`
- [关系抽取](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/relation_extractor.py): `GuJi古籍翻译/src/relation_extractor.py`
- [知识图谱构建](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/GuJi%E5%8F%A4%E7%B1%8D%E7%BF%BB%E8%AF%91/src/knowledge_graph.py): `GuJi古籍翻译/src/knowledge_graph.py`

### 个人知识库 RAG

- [RAG引擎](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/engine.py): `RAG/rag/engine.py`
- [索引构建](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/pipeline.py): `RAG/rag/ingestion/pipeline.py`
- [文档加载](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/loaders.py): `RAG/rag/ingestion/loaders.py`
- [文本分片](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/ingestion/chunking.py): `RAG/rag/ingestion/chunking.py`
- [向量存储](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/vector_store.py): `RAG/rag/retrieval/vector_store.py`
- [文本嵌入](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/embeddings.py): `RAG/rag/retrieval/embeddings.py`
- [BM25索引](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/bm25.py): `RAG/rag/retrieval/bm25.py`
- [混合检索](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/hybrid.py): `RAG/rag/retrieval/hybrid.py`
- [结果重排](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/retrieval/reranker.py): `RAG/rag/retrieval/reranker.py`
- [上下文与提示](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/generation/prompts.py): `RAG/rag/generation/prompts.py`
- [LLM客户端](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/RAG/rag/generation/llm.py): `RAG/rag/generation/llm.py`

### 停车管理

- [前端应用](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/%E5%81%9C%E8%BD%A6%E7%AE%A1%E7%90%86%E7%B3%BB%E7%BB%9F/frontend/js/app.js): `停车管理系统/frontend/js/app.js`
- [后端应用](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/%E5%81%9C%E8%BD%A6%E7%AE%A1%E7%90%86%E7%B3%BB%E7%BB%9F/backend/app.py): `停车管理系统/backend/app.py`

### AI介绍网站

- [网站主脚本](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/AI%E6%8A%80%E6%9C%AF%E4%BB%8B%E7%BB%8D%E7%BD%91%E9%A1%B5/js/main.js): `AI技术介绍网页/js/main.js`
- [页面组件](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/AI%E6%8A%80%E6%9C%AF%E4%BB%8B%E7%BB%8D%E7%BD%91%E9%A1%B5/js/components.js): `AI技术介绍网页/js/components.js`
- [AI聊天界面](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/AI%E6%8A%80%E6%9C%AF%E4%BB%8B%E7%BB%8D%E7%BD%91%E9%A1%B5/js/ai-chat.js): `AI技术介绍网页/js/ai-chat.js`

### CNN视觉识别

- [CNN识别模型](https://github.com/yh7dfhvbk4-code/ryy-projects/blob/main/cnn%20%E6%96%91%E9%A9%AC%E7%BA%BF%E8%AF%86%E5%88%AB/cnn%E6%A8%A1%E5%9E%8B%E9%9A%9C%E7%A2%8D%E7%89%A9%E8%AF%86%E5%88%AB.py): `cnn 斑马线识别/cnn模型障碍物识别.py`

### Other components

- **文本用户**
- **Neo4j图数据库**
- **知识库用户**
- **兼容LLM服务**
- **停车用户**
- **网站访客**
- **识别使用者**

## Connections

- 文本用户 → Web API: 提交文本
- Web API → 文本流水线: 调用流水线 (evidence: `GuJi古籍翻译/src/web_api.py`)
- 文本流水线 → 文本清洗: 清洗文本 (evidence: `GuJi古籍翻译/src/pipeline.py`)
- 文本流水线 → 分词断句: 分词断句 (evidence: `GuJi古籍翻译/src/pipeline.py`)
- 文本流水线 → 实体识别: 识别实体 (evidence: `GuJi古籍翻译/src/pipeline.py`)
- 实体识别 → BERT-CRF模型: 调用模型 (evidence: `GuJi古籍翻译/src/ner/predictor.py`)
- 文本流水线 → 关系抽取: 抽取关系 (evidence: `GuJi古籍翻译/src/pipeline.py`)
- 文本流水线 → 知识图谱构建: 构建图谱 (evidence: `GuJi古籍翻译/src/pipeline.py`)
- 知识图谱构建 → Neo4j图数据库: 写入图数据
- NER数据集 → BERT-CRF模型: 提供训练数据
- 少样本处理 → NER数据集: 增强样本 (evidence: `GuJi古籍翻译/src/few_shot.py`)
- 知识库用户 → RAG引擎: 提问建库
- RAG引擎 → 索引构建: 构建索引 (evidence: `RAG/rag/engine.py`)
- 索引构建 → 文档加载: 加载文档 (evidence: `RAG/rag/ingestion/pipeline.py`)
- 索引构建 → 文本分片: 切分文档 (evidence: `RAG/rag/ingestion/pipeline.py`)
- 索引构建 → 向量存储: 写入分片 (evidence: `RAG/rag/ingestion/pipeline.py`)
- 索引构建 → BM25索引: 重建索引 (evidence: `RAG/rag/ingestion/pipeline.py`)
- 向量存储 → 文本嵌入: 向量编码 (evidence: `RAG/rag/retrieval/vector_store.py`)
- RAG引擎 → BM25索引: 加载索引 (evidence: `RAG/rag/engine.py`)
- RAG引擎 → 混合检索: 混合召回 (evidence: `RAG/rag/engine.py`)
- 混合检索 → 向量存储: 向量检索 (evidence: `RAG/rag/retrieval/hybrid.py`)
- 混合检索 → BM25索引: 关键词检索 (evidence: `RAG/rag/retrieval/hybrid.py`)
- RAG引擎 → 结果重排: 重排候选 (evidence: `RAG/rag/engine.py`)
- RAG引擎 → 上下文与提示: 组装上下文 (evidence: `RAG/rag/engine.py`)
- RAG引擎 → LLM客户端: 生成回答 (evidence: `RAG/rag/engine.py`)
- LLM客户端 → 兼容LLM服务: 请求补全 (evidence: `RAG/rag/generation/llm.py`)
- 停车用户 → 前端应用: 使用界面
- 前端应用 → 后端应用: 请求服务
- 网站访客 → 网站主脚本: 访问网站
- 网站访客 → AI聊天界面: 使用聊天
- 识别使用者 → CNN识别模型: 提交图像
