This project demonstrates the use of LangChain, a Python library for building applications with language models.
To get started with this project, you need to install several packages. You can do this by running the following commands in your terminal:
pip install -U langchain==0.2.0
pip install -U langchain-core==0.2.0
pip install -U langchain-community==0.2.0
pip install -U langchain-openai==0.1.7
pip install -U langchain-experimental==0.0.59
pip install -U wikipedia
pip install -U yfinance
pip install -U tabulate
pip install python-dotenv
This project contains a Jupyter notebook demo.ipynb
that demonstrates various features of LangChain, including:
- Prompt Templates: Predefined recipes for generating prompts for language models.
- Output Parsers: Transform the output of a language model to a more suitable format.
- Chaining: Execute a sequence of steps or operations in a specific order to accomplish a task.
- Memory: Refer to information introduced earlier in the conversation.
- Tools and Toolkits: Interfaces that an agent, chain, or language model can use to interact with the world.
- Agents: Use a language model to choose a sequence of actions to take.
To run the demo, open the demo.ipynb notebook in Jupyter Notebook and run the cells in order.