Langchain Projects

This repository contains experiments and sample projects using LangChain for building LLM-powered applications, including Retrieval-Augmented Generation (RAG) pipelines.

Features

• Chat with OpenAI models using LangChain's ChatOpenAI
• Document loading and splitting (PDFs, text)
• Embeddings generation with OpenAI models
• Vector storage and retrieval using Pinecone
• Retrieval-Augmented Generation (RAG) with LangChain's RetrievalQA chain

Setup

1. Clone the repository:

   git clone https://github.com/your-username/langchain-projects.git
   cd langchain-projects

2. Install dependencies:

   poetry install
   # or, if using pip:
   pip install -r requirements.txt

3. Set environment variables:

   • OPENAI_API_KEY for OpenAI access
   • PINECONE_API_KEY for Pinecone access

4. Activate virtual environment:

   poetry shell

Usage

• Run Jupyter notebooks for interactive experiments:

  jupyter notebook

• Example notebooks:
  • chat_model.ipynb: Basic chat with OpenAI models
  • rag.ipynb: Full RAG pipeline with PDF ingestion, chunking, embedding, and retrieval

Notes

• Adjust chunk sizes in text splitters to avoid Pinecone's per-vector and per-request size limits.
• Ensure your embedding model's output dimension matches your Pinecone index dimension.