# eniolaademola/langchain-basics architecture

> Architecture diagram of the GitHub repository eniolaademola/langchain-basics, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-05-26). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/eniolaademola/langchain-basics
- Repository: https://github.com/eniolaademola/langchain-basics

## Overview

##### Project type and purpose
- This is an educational/sample repository for demonstrating LangChain fundamentals in Python using Jupyter notebooks.
- It walks through a retrieval-augmented generation (RAG) pipeline: ingesting data, transforming it, creating embeddings, storing vectors, and running queries against a language model.

##### File structure
- Top-level directories are numbered to reflect pipeline stages:
- 0.0-basics – core LangChain/LLM interaction examples (OpenAI & Ollama)
- 1.0-data_ingestion – document loaders and data-loading demos
- 2.0-data-transformation – text splitting strategies
- 3.0-embeddings – embedding generation via OpenAI, Ollama, Hugging Face
- 4.0-vector_store_db – vector database demos using FAISS and ChromaDB
- Each directory contains Jupyter notebooks (.ipynb) and sample data files.
- Root contains README.md, requirements.txt, and .gitignore.

##### README for additional insights
- Confirms the repo is a step-by-step tutorial covering:
- Foundational LangChain concepts
- Data ingestion/transformation pipelines
- Embedding creation and comparison
- Vector store setup (FAISS, ChromaDB)

##### Elements
- Clear, concise labels for each component and technology
- Blue (data sources/files)
- Green (processing stages: ingestion, transformation, embeddings)
- Orange (storage: vector DBs)
- Purple (LLM/query/execution)

## Diagram (Mermaid)

```mermaid
flowchart TD
    %% Data Sources
    DataSources["Data Sources\n(local PDF, HTML, text)"]:::data
    click DataSources "https://github.com/eniolaademola/langchain-basics/tree/main/1.0-data_ingestion/data/"

    %% Ingestion Layer
    IngestionLayer["Ingestion Layer\n(LangChain Document Loaders)"]:::process
    click IngestionLayer "https://github.com/eniolaademola/langchain-basics/blob/main/1.0-data_ingestion/1.1-data_ingestion.ipynb"

    %% Transformation Layer
    subgraph "Transformation Layer\n(Text Splitters)" 
        direction TB
        RecSplit["Recursive Character Splitter"]:::process
        click RecSplit "https://github.com/eniolaademola/langchain-basics/blob/main/2.0-data-transformation/2.1-recursive_character_text_splitter.ipynb"
        CharSplit["Character Text Splitter"]:::process
        click CharSplit "https://github.com/eniolaademola/langchain-basics/blob/main/2.0-data-transformation/2.2-character_text_splitter.ipynb"
        HTMLSplit["HTML Text Splitter"]:::process
        click HTMLSplit "https://github.com/eniolaademola/langchain-basics/blob/main/2.0-data-transformation/2.3-HTML_text_splitter.ipynb"
        JSONSplit["Recursive JSON Splitter"]:::process
        click JSONSplit "https://github.com/eniolaademola/langchain-basics/blob/main/2.0-data-transformation/2.4-recursive_Json_Splitter.ipynb"
    end

    %% Embedding Layer
    subgraph "Embedding Layer"
        direction TB
        OpenAIEmb["OpenAI Embeddings"]:::process
        click OpenAIEmb "https://github.com/eniolaademola/langchain-basics/blob/main/3.0-embeddings/3.1-openai-embedding.ipynb"
        OllamaEmb["Ollama Embeddings"]:::process
        click OllamaEmb "https://github.com/eniolaademola/langchain-basics/blob/main/3.0-embeddings/3.2-ollama_embedding.ipynb"
        HFEmb["Hugging Face Embeddings"]:::process
        click HFEmb "https://github.com/eniolaademola/langchain-basics/blob/main/3.0-embeddings/3.3-huggingface_embedding.ipynb"
    end

    %% Vector Store
    subgraph "Vector Store"
        direction TB
        FAISSNode["FAISS Index"]:::storage
        click FAISSNode "https://github.com/eniolaademola/langchain-basics/blob/main/4.0-vector_store_db/4.1-faiss.ipynb"
        FAISSFiles["FAISS Index Files"]:::storage
        click FAISSFiles "https://github.com/eniolaademola/langchain-basics/tree/main/4.0-vector_store_db/faiss_db/"
        ChromaNode["ChromaDB Service"]:::storage
        click ChromaNode "https://github.com/eniolaademola/langchain-basics/blob/main/4.0-vector_store_db/4.2-chromadb.ipynb"
        ChromaFiles["ChromaDB Files"]:::storage
        click ChromaFiles "https://github.com/eniolaademola/langchain-basics/tree/main/4.0-vector_store_db/chroma_db/"
    end

    %% Query/LLM Layer
    subgraph "Query / LLM Layer"
        direction TB
        RAGChain["Core LangChain RAG Examples"]:::query
        click RAGChain "https://github.com/eniolaademola/langchain-basics/blob/main/0.0-basics/getting_started.ipynb"
        OllamaChat["Ollama Chat Script"]:::query
        click OllamaChat "https://github.com/eniolaademola/langchain-basics/blob/main/0.0-ollama/chat_with_ollama_.py"
    end

    %% User Environment
    User["User / Notebook Environment\n(Jupyter)"]:::query

    %% Connections
    DataSources -->|loads documents| IngestionLayer
    IngestionLayer -->|splits text| RecSplit
    RecSplit --> CharSplit
    CharSplit --> HTMLSplit
    HTMLSplit --> JSONSplit
    JSONSplit -->|documents chunks| OpenAIEmb
    JSONSplit -->|documents chunks| OllamaEmb
    JSONSplit -->|documents chunks| HFEmb
    OpenAIEmb -->|vectors| FAISSNode
    OpenAIEmb -->|vectors| ChromaNode
    OllamaEmb -->|vectors| FAISSNode
    OllamaEmb -->|vectors| ChromaNode
    HFEmb -->|vectors| FAISSNode
    HFEmb -->|vectors| ChromaNode
    FAISSNode -->|persist/index| FAISSFiles
    ChromaNode -->|persist/index| ChromaFiles
    User -->|query| RAGChain
    User -->|query| OllamaChat
    RAGChain -->|similarity search| FAISSNode
    RAGChain -->|similarity search| ChromaNode
    RAGChain -->|calls LLM| OpenAIEmb
    OllamaChat -->|similarity search| FAISSNode
    OllamaChat -->|similarity search| ChromaNode
    OllamaChat -->|calls LLM| OllamaEmb
    RAGChain -->|response| User
    OllamaChat -->|response| User

    %% Styles
    classDef data fill:#D0E8FF,stroke:#0080FF,color:#003366;
    classDef process fill:#D6FFD0,stroke:#33AA00,color:#003300;
    classDef storage fill:#FFE8B0,stroke:#FF9900,color:#663300;
    classDef query fill:#E8D0FF,stroke:#8000B3,color:#330033;
```
