CodexploreRepo/langchain

LangChain

Introduction

• LangChain Introduction
• Prompt Templates
• Output Parsers

RAG

• Indexing
  • Document Loading
  • Document Splitting
  • Embeddings & Vector Stores

  
  RAG Indexing: Load, Split, Embed, Store

• Retrieval
  • RetrivalQA Chain: if the retrieved documents cannot fit into the context window, we can use map_reduce, refine, and map_rerank to solve this problem
    • Limitation of RetrivalQA: fails to preserve conversational history as RetrievalQA chain does not have any concept of state
  • Conversational Retrieval Chain = Conversation memory + RetrievalQA Chain

Given a user input, relevant splits are retrieved from storage using a Retriever & ChatModel / LLM produces an answer using a prompt that includes the question and the retrieved data

Tools

• LangSmith

Installation

• Core LangChain: pip install --upgrade langchain

Resources

• Deep Learning's LangChain Short Course