https://github.com/developersdigest/get_started_with_langchain_in_nodejs

https://github.com/developersdigest/get_started_with_langchain_in_nodejs

Last synced: 8 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/developersdigest/get_started_with_langchain_in_nodejs
• Owner: developersdigest
• Created: 2023-04-20T14:16:36.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2024-09-25T20:54:46.000Z (over 1 year ago)
• Last Synced: 2025-05-15T04:33:26.853Z (10 months ago)
• Language: JavaScript
• Size: 14.6 KB
• Stars: 44
• Watchers: 3
• Forks: 14
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          

# Retrieval-based QA System with Embeddings using LangChain

This project implements a retrieval-based question answering (QA) system using embeddings and the LangChain library. The system processes a text file, creates a vector store, and uses it to answer questions based on the content.

## Contents

The project consists of the following key files:

- `index.js`: The main JavaScript file containing the core logic

- `.env`: Environment file for storing API keys

- `package.json` and `package-lock.json`: Node.js project configuration files

- `readme.md`: Project documentation

## How it Works

1. **Setup and Initialization**:

   - The code imports necessary modules from LangChain and other libraries.

   - It loads environment variables, including the OpenAI API key.

2. **Vector Store Creation**:

   - If a vector store doesn't exist, the system:

     - Reads the input text file

     - Splits the text into chunks

     - Creates embeddings using OpenAI's model

     - Stores these embeddings in a vector store (HNSWLib)

3. **Question Answering**:

   - The system uses a RetrievalQAChain, which combines:

     - The OpenAI language model

     - The vector store as a retriever

4. **Query Processing**:

   - When given a question, the system:

     - Retrieves relevant information from the vector store

     - Uses the language model to generate an answer based on the retrieved context

## Usage

To use this system:

1. Ensure you have the required dependencies installed (see `package.json`).

2. Set up your OpenAI API key in the `.env` file.

3. Prepare your input text file (default name: "The_Creative_Act").

4. Run the `index.js` script, which will:

   - Create or load the vector store

   - Process the specified question

   - Output the answer

## Input and Output

### Input:

- A text file containing the corpus of information

- A question string defined in the `index.js` file

### Output:

- The system logs the answer to the console

## Dependencies

The project relies on several npm packages, including:

- `langchain`: For the core QA and embedding functionality

- `hnswlib-node`: For efficient similarity search

- `dotenv`: For managing environment variables

## Notes

- The system uses OpenAI's API, so ensure you have proper authentication set up.

- The vector store is saved locally, allowing for faster subsequent runs.

- This implementation is designed for Node.js environments.