Efficiently use Langchain for Complex Tasks

🚀 Installation

Pre-Setup

Prerequisites:

• Python
• Git
• OpenAI API Key

Setup

git clone https://github.com/Haste171/langchain-chatbot.git

Recommended to use a virtual environment

pip install -r requirements.txt

Reference example.env to create credentials file

OPENAI_API_KEY=
PINECONE_API_KEY=
PINECONE_ENVIRONMENT=
PINECONE_INDEX=

Run Terminal Interface

python chatbot.py

Run Chat Interface

streamlit run streamlit.py

In-App Usage

Click to expand

Terminal

For usage of the terminal interface place files in the docs folder to be ingested

Once Files are ingested one can choose to ingest more files in future usage or just query the existing vector database

Chat Interface

For usage of the chat interface upload files directly to the Browse Files section

All

• Temperature: The amount of creativity/burstiness the AI will use when querying files
• Sources: The amount of sources the AI will base it's answer off of and use for context

🔧 Key Features

• Conversational answers with chat history
• Compatibility for PDF documents (more soon)
• Local and external vector database compatibility

Soon:

• Externally hosted user accessible chat interface
• Compatibility with multiple files types (Llama Index)
• Compatibility with offline models (HuggingFace, Vicuna, Alpaca)

💻 Contributing

If you would like to contribute to the LangChain Chatbot, please follow these steps:

1. Fork the repository
2. Create a new branch for your feature or bug fix
3. Write tests for your changes
4. Implement your changes and ensure that all tests pass
5. Submit a pull request

📝 Credits

The LangChain Chatbot was developed by Haste171 with much inspiration from Mayo with the GPT4 & LangChain Chatbot for large PDF docs. This project is mainly a port to Python from the Mayo chatbot.

🔨 License

The LangChain Chatbot is released under the MIT License.

💻 Terminal

💻 Interface