https://github.com/danitilahun/langchain_projects

This repository hosts diverse Langchain projects featuring chains, retrievers, tools, agents, RAG pipelines, vector databases (FAISS, Pinecone , Chroma), and integrations with Langsmith and Langserve via FastAPI.
https://github.com/danitilahun/langchain_projects

chroma embeddings faiss-vector-database fastapi langchain llm ollama openai pinecone rag retrieval-augmented-generation vector-database

Last synced: 4 months ago
JSON representation

This repository hosts diverse Langchain projects featuring chains, retrievers, tools, agents, RAG pipelines, vector databases (FAISS, Pinecone , Chroma), and integrations with Langsmith and Langserve via FastAPI.

• Host: GitHub
• URL: https://github.com/danitilahun/langchain_projects
• Owner: Danitilahun
• Created: 2025-02-23T08:33:22.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2025-02-23T14:55:02.000Z (4 months ago)
• Last Synced: 2025-02-23T15:35:00.741Z (4 months ago)
• Topics: chroma, embeddings, faiss-vector-database, fastapi, langchain, llm, ollama, openai, pinecone, rag, retrieval-augmented-generation, vector-database
• Language: Jupyter Notebook
• Homepage:
• Size: 5.31 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
Langchain Projects

## Visual Overviews

### General Ecosystem

![Langchain Overview](image/langchain_stack.webp)  

*This image provides a high-level view of Langchain's ecosystem, showcasing its modular design and the integration of various components such as chains, agents, retrievers, and vector databases.*

### Architecture Diagram

![Langchain Architecture](image/modules.png)  

*This diagram highlights key components of Langchain, including:  

	- **Chains & Agents:** Modules to manage sequential tasks and decision-making.  

	- **Retrievers & RAG:** Systems to fetch and augment data for improved responses.  

	- **Vector Databases:** Integration with FAISS, Chroma, and more for efficient data storage and retrieval.  

	- **LLMs & Embeddings:** Support for models like OpenAI and Ollama, with embedding techniques from Hugging Face and others.  

	- **Langsmith & Langserve:** Tools to monitor, serve, and scale applications using FastAPI.*

### RAG Pipeline Overview

![RAG Pipeline](image/rag-pipeline.webp)  

*This image illustrates the Retrieval-Augmented Generation (RAG) pipeline. It outlines how external data is retrieved and combined with language model outputs to improve response quality. The pipeline typically involves:  

	- **Data Retrieval:** Searching external databases for relevant context.  

	- **Integration:** Merging retrieved data with generated content.  

	- **Generation:** Producing enriched outputs that leverage both internal and external information.*

### Hybrid Search Overview

![Hybrid Search](image/hybridsearch.png)  

*This image demonstrates the hybrid search mechanism, which combines dense (vector-based) and sparse (keyword-based) retrieval techniques. It employs Reciprocal Rank Fusion (RRF) to integrate results from multiple search methods, assigning higher scores to consistently high-ranking items for improved overall search accuracy.*

## About This Repository

This repository holds a variety of projects related to Langchain, including:

- **Chains, Retrievers, Tools & Agents:** Build and integrate dynamic components.

- **Retrieval-Augmented Generation (RAG):** Enhance outputs using external data sources.

- **Vector Databases:** Utilize FAISS, Pinecone, Chroma, and similar tools for efficient storage and retrieval.

- **Langsmith & Langserve:** Develop and serve applications with integrated monitoring and FastAPI support.

- **Large Language Models:** Experiment with OpenAI, Ollama models, and more.

- **Word Embeddings:** Implement diverse embedding techniques from Hugging Face, OpenAI, Ollama, etc.