# pacificrm/rag-architecture architecture

> Architecture diagram of the GitHub repository pacificrm/rag-architecture, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-07-17). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/pacificrm/rag-architecture
- Repository: https://github.com/pacificrm/rag-architecture

## Overview

##### Project Type and Purpose
- This is a full-stack Python Streamlit application that supports multimodal document ingestion (audio, video, image, text, CSV, PDF, DOCX, JSON, YAML) and provides a natural-language query interface via a Retrieval-Augmented Generation (RAG) pipeline.
- The purpose is to extract text from files, store vector embeddings in a Milvus database, and answer queries by retrieving relevant embeddings and generating responses via an LLM.

##### File Structure
- Top-level files:
- app.py – Main Streamlit application (UI + controller).
- requirements.txt – Python dependencies.
- milvus_database.db – Local Milvus vector database file.
- Dataset/Images/ – Raw data storage.
- utils/ – Collection of processing modules:
– audio_utils.py, video_utils.py, image_utils.py, document_loaders.py, milvus_client.py.
- No separate “frontend” folder: Streamlit handles UI and server logic in app.py.
- Configs: requirements.txt; no Docker/Kubernetes scripts.

##### README for Additional Insights
- Dependencies: LangChain, HuggingFaceEmbeddings, Milvus, EasyOCR, speech_recognition, pydub, MoviePy, transformers.
- Features: upload & process, embed & store, query & generate.

##### Main Components and Relationships
a. Components
1. Client Browser – User interacts with Streamlit UI.
2. Streamlit App (app.py)
- Upload Mode: accepts file → routes to appropriate utils module.
- Query Mode: accepts question → triggers RAG pipeline.
3. Utility Modules (utils/)
- audio_utils, video_utils, image_utils, document_loaders: extract raw text.
- milvus_client: connect, insert, search embeddings.
4. Embedding Service
- LangChain + HuggingFaceEmbeddings for vectorization.
5. Milvus Database
- Stores and retrieves embeddings.
6. LLM / RAG Pipeline
- LangChain chains for retrieval and generation.
7. External ML Models/Services
- EasyOCR, speech_recognition, pydub, MoviePy, transformer models from Hugging Face.
8. Local Storage
- Dataset/ folder, Images/ folder, milvus_database.db.

b. Data Flows and Interactions
- Upload Flow:
Browser → Streamlit UI → app.py → utils module → raw text → embeddings → milvus_client → Milvus → confirmation back to UI.
- Query Flow:
Browser → UI → app.py → milvus_client search → retrieved passages → LLM (RAG) → generated answer → UI.
- External Models invoked by utils modules and embedding service.

c. Architectural Patterns and Principles
- Separation of Concerns: UI vs. processing vs. storage.
- Modular design: one utils module per modality.
- Retrieval-Augmented Generation: vector DB + LLM chain.
- Stateful database for embeddings; stateless processing modules.

d. Key Technologies / Frameworks
- Streamlit (UI + server)
- LangChain (orchestration, embedding wrapper)
- HuggingFaceEmbeddings (vector creation)
- Milvus (vector database)
- EasyOCR, speech_recognition, pydub, MoviePy (data extraction)
- Python 3.8+, various LangChain document loaders

## Diagram (Mermaid)

```mermaid
flowchart TB
    %% UI Layer
    subgraph "UI Layer"
        direction TB
        Browser["Client Browser"]:::ui
        StreamlitApp["Streamlit App (app.py)"]:::ui
    end

    %% Processing Layer
    subgraph "Processing Modules" 
        direction TB
        AudioUtils["audio_utils.py"]:::processing
        VideoUtils["video_utils.py"]:::processing
        ImageUtils["image_utils.py"]:::processing
        DocLoaders["document_loaders.py"]:::processing
        MilvusClientModule["milvus_client.py"]:::processing
    end

    %% Embedding Layer
    subgraph "Embedding Layer"
        direction TB
        HuggingFaceEmb["HuggingFaceEmbeddings"]:::emb
        MilvusClient["Milvus Client"]:::emb
    end

    %% Database Layer
    subgraph "Database Layer"
        direction TB
        MilvusDB["Milvus Database\n(milvus_database.db)"]:::db
    end

    %% RAG Pipeline
    subgraph "RAG Pipeline"
        direction TB
        VectorSearch["Vector Search"]:::rag
        LLMChain["LLM Chain (LangChain RAG)"]:::rag
    end

    %% External Models
    subgraph "External Models/Services"
        direction TB
        EasyOCR["EasyOCR"]:::external
        SpeechRec["speech_recognition"]:::external
        Pydub["pydub"]:::external
        MoviePy["MoviePy"]:::external
        Transformers["Hugging Face Models"]:::external
    end

    %% File Storage
    subgraph "Local Storage"
        direction TB
        Dataset["Dataset/"]:::storage
        Images["Images/"]:::storage
    end

    %% Dependency & Metadata
    Requirements["requirements.txt"]:::meta
    TestHarness["test.py"]:::meta
    README["README.md"]:::meta

    %% Flows
    Browser -->|select file / ask query| StreamlitApp
    StreamlitApp -->|upload file| AudioUtils
    StreamlitApp -->|upload file| VideoUtils
    StreamlitApp -->|upload file| ImageUtils
    StreamlitApp -->|upload file| DocLoaders
    AudioUtils -->|extract text| HuggingFaceEmb
    VideoUtils -->|extract text| HuggingFaceEmb
    ImageUtils -->|extract text| HuggingFaceEmb
    DocLoaders -->|load text| HuggingFaceEmb

    HuggingFaceEmb -->|create embeddings| MilvusClient
    MilvusClient -->|insert/search vectors| MilvusDB

    Browser -->|query| StreamlitApp
    StreamlitApp -->|search vectors| MilvusClient
    MilvusClient -->|retrieved passages| VectorSearch
    VectorSearch -->|provide context| LLMChain
    LLMChain -->|generate answer| StreamlitApp

    %% External invocations
    AudioUtils -->|uses| SpeechRec
    AudioUtils -->|uses| Pydub
    VideoUtils -->|uses| MoviePy
    ImageUtils -->|uses| EasyOCR
    HuggingFaceEmb -->|uses| Transformers

    %% Metadata links
    StreamlitApp --> Requirements
    HuggingFaceEmb --> Requirements
    AudioUtils --> Requirements
    VideoUtils --> Requirements
    ImageUtils --> Requirements
    DocLoaders --> Requirements
    MilvusClientModule --> Requirements

    %% Click Events
    click StreamlitApp "https://github.com/pacificrm/rag-architecture/blob/main/app.py"
    click AudioUtils "https://github.com/pacificrm/rag-architecture/blob/main/utils/audio_utils.py"
    click VideoUtils "https://github.com/pacificrm/rag-architecture/blob/main/utils/video_utils.py"
    click ImageUtils "https://github.com/pacificrm/rag-architecture/blob/main/utils/image_utils.py"
    click DocLoaders "https://github.com/pacificrm/rag-architecture/blob/main/utils/document_loaders.py"
    click MilvusClientModule "https://github.com/pacificrm/rag-architecture/blob/main/utils/milvus_client.py"
    click Requirements "https://github.com/pacificrm/rag-architecture/blob/main/requirements.txt"
    click MilvusDB "https://github.com/pacificrm/rag-architecture/blob/main/milvus_database.db"
    click Dataset "https://github.com/pacificrm/rag-architecture/tree/main/Dataset/"
    click Images "https://github.com/pacificrm/rag-architecture/tree/main/Images/"
    click TestHarness "https://github.com/pacificrm/rag-architecture/blob/main/test.py"
    click README "https://github.com/pacificrm/rag-architecture/blob/main/README.md"

    %% Styles
    classDef ui fill:#D0E6FF,stroke:#0366D6,stroke-width:1px;
    classDef processing fill:#E6F4EA,stroke:#22863A,stroke-width:1px;
    classDef emb fill:#FFF5B1,stroke:#DBAB09,stroke-width:1px;
    classDef db fill:#FFEFEF,stroke:#D73A49,stroke-width:1px;
    classDef rag fill:#F0F0F0,stroke:#6F42C1,stroke-width:1px;
    classDef external fill:#F5F5F5,stroke:#0366D6,stroke-width:1px,stroke-dasharray: 5 5;
    classDef storage fill:#EAEAEA,stroke:#6A737D,stroke-width:1px;
    classDef meta fill:#FFFFFF,stroke:#CCCCCC,stroke-width:1px,stroke-dasharray: 2 2;
```
