LangChain & OpenAI Example Project

Overview

This project is a simple example demonstrating the use of LangChain and OpenAI for retrieving and generating text-based answers based on a set of documents. The project primarily utilizes OpenAI's GPT-3.5-turbo model, along with a document retrieval system powered by Chroma and OpenAI embeddings.

Features

• Document Embeddings: Uses OpenAI embeddings to convert documents into vectors for similarity search.
• Retrieval System: Employs Chroma to perform similarity searches against embedded documents.
• Language Model: Utilizes OpenAI's GPT-3.5-turbo for generating responses based on retrieved documents.
• Prompting System: Uses LangChain's template system for structuring prompts to the language model.

Setup and Usage

Installation

Make sure to install the necessary Python packages. You can typically do this with pip:

pip install langchain langchain-extras openai

Environment Variables

Ensure you have a .env file or set the necessary environment variables for OpenAI and other services you might be using.

Running the Script

Execute the main script to see the output for a given sample question:

python main.py

The script is designed to answer the question "tell me about cats" using the provided set of documents.

License

This project is licensed under the MIT License.