LangChain Chat and Chroma Embeddings Integration

Overview

tchat is a Python project with two parts: a chat application built on LangChain and a system for document retrieval and question answering using Chroma embeddings.

LangChain Chat Application

Features

• OpenAI Chat Model: Uses ChatOpenAI for language model interactions.
• Conversation Memory: Incorporates ConversationSummaryMemory for conversation history.
• Dynamic Prompting: Implements ChatPromptTemplate and HumanMessagePromptTemplate for context-aware prompts.
• Environment Management: Uses dotenv for configuration.

How It Works

1. Environment Setup: Load necessary variables.
2. Model Initialization: Set up OpenAI chat model.
3. Memory Management: Store and recall conversation history.
4. Prompt Configuration: Customize prompts.
5. Interactive Chat: Continuously generate context-based responses.

Usage

1. Install Dependencies (dotenv, langchain).
2. Clone Repository.
3. Set Up Environment (API keys, etc.).
4. Run Main Script.

Chroma Embeddings for Text Files

Features

• Document Processing: Uses TextLoader and CharacterTextSplitter.
• Embedding Generation: Uses OpenAIEmbeddings.
• Chroma Vector Store: Manages document embeddings.
• Question-Answering: Uses RetrievalQA and ChatOpenAI.

How It Works

• main.py: Initializes components, loads documents, creates Chroma database, and performs similarity searches.
• prompt.py: Sets up ChatOpenAI model, integrates Chroma, and runs RetrievalQA chain.

Usage

1. Prepare Environment.
2. Clone Repository.
3. Load facts.txt.
4. Run main.py and prompt.py for queries.

Visit the project on GitHub for the code and more details.