chatformers 1.0.6

Chatformers is a Python package that simplifies chatbot development by automatically managing chat history using local vector databases like Chroma DB.

Chatformers

⚡ Chatformers is a Python package designed to simplify the development of chatbot applications that use Large Language Models (LLMs). It offers automatic chat history management using a local vector database (ChromaDB, Qdrant or Pgvector), ensuring efficient context retrieval for ongoing conversations.

Install

pip install chatformers

Documentation-

https://coda.io/@chatformers/chatformers

Why Choose chatformers?

1. Effortless History Management: No need to manage extensive chat history manually; the package automatically handles it.
2. Simple Integration: Build a chatbot with just a few lines of code.
3. Full Customization: Maintain complete control over your data and conversations.
4. Framework Compatibility: Easily integrate with any existing framework or codebase.

Key Features

1. Easy Chatbot Creation: Set up a chatbot with minimal code.
2. Automated History Management: Automatically stores and fetches chat history for context-aware conversations.

How It Works

1. Project Setup: Create a basic project structure.
2. Automatic Storage: Chatformers stores your conversations (user inputs and AI outputs) in VectorDB.
3. Contextual Conversations: The chatbot fetches relevant chat history whenever you engage with the LLM.

Prerequisites-

1. Python: Ensure Python is installed on your system.
2. GenAI Knowledge: Familiarity with Generative AI models.

Example Usage-

Read Documentation for advanced usage and understanding: https://coda.io/@chatformers/chatformers

    from chatformers.chatbot import Chatbot
    import os
    from openai import OpenAI

    system_prompt = None  # use the default
    metadata = None  # use the default metadata
    user_id = "Sam-Julia"
    chat_model_name = "llama-3.1-8b-instant"
    memory_model_name = "llama-3.1-8b-instant"
    max_tokens = 150  # len of tokens to generate from LLM
    limit = 4  # maximum number of memory to added during LLM chat
    debug = True  # enable to print debug messages

    os.environ["GROQ_API_KEY"] = ""
    llm_client = OpenAI(base_url="https://api.groq.com/openai/v1",
                        api_key="",
                        )  # Any OpenAI Compatible LLM Client, using groq here
    config = {
        "vector_store": {
            "provider": "chroma",
            "config": {
                "collection_name": user_id,
                "path": "db",
            }
        },
        "embedder": {
            "provider": "ollama",
            "config": {
                "model": "nomic-embed-text:latest"
            }
        },
        "llm": {
            "provider": "groq",
            "config": {
                "model": memory_model_name,
                "temperature": 0.1,
                "max_tokens": 1000,
            }
        },
    }

    chatbot = Chatbot(config=config, llm_client=llm_client, metadata=None, system_prompt=system_prompt,
                      chat_model_name=chat_model_name, memory_model_name=memory_model_name,
                      max_tokens=max_tokens, limit=limit, debug=debug)

    # Example to add buffer memory
    memory_messages = [
        {"role": "user", "content": "My name is Sam, what about you?"},
        {"role": "assistant", "content": "Hello Sam! I'm Julia."}
    ]
    chatbot.add_memories(memory_messages, user_id=user_id)

    # Buffer window memory, this will be acts as sliding window memory for LLM
    message_history = [{"role": "user", "content": "where r u from?"},
                       {"role": "assistant", "content": "I am from CA, USA"}]

    # Example to chat with the bot, send latest / current query here
    query = "Do you remember my name?"
    response = chatbot.chat(query=query, message_history=message_history, user_id=user_id, print_stream=True)
    print("Assistant: ", response)

    # Example to check memories in bot based on user_id
    # memories = chatbot.get_memories(user_id=user_id)
    # for m in memories:
    #     print(m)
    # print("================================================================")
    # related_memories = chatbot.related_memory(user_id=user_id,
    #                                           query="yes i am sam? what us your name")
    # print(related_memories)

FAQs-

1. Can I customize LLM endpoints / Groq or other models?

   • Yes, any OpenAI-compatible endpoints and models can be used.

2. Can I use custom hosted chromadb

   • Yes, you can specify custom endpoints for Chroma DB. If not provided, a Chroma directory will be created in your project's root folder.

3. I don't want to manage history. Just wanted to chat.

   • Yes, set memory=False to disable history management and chat directly.

4. Need help or have suggestions?

   • Raise an issue or contact me at [e-mail]

Star History