This project leverages the LangChain and CrewAI frameworks to create an AI system composed of specialized agents working collaboratively to perform complex tasks such as statistical/causal analysis and decision-making. By integrating semantic analysis, data preprocessing, statistical modeling, and decision analysis, our system aims to provide comprehensive insights and actionable recommendations from diverse datasets.
- Modular AI Agents: Individual agents dedicated to specific tasks like data preprocessing, statistical analysis, causal inference, decision analysis, and semantic analysis.
- Collaborative Workflow: Utilizes CrewAI to enable agents to work together as a team, sharing information and tasks to achieve complex objectives efficiently.
- Incremental Knowledge Base Updating: Supports continuous learning and updating of the system's knowledge base, ensuring decisions are based on the latest available information.
- Custom Tools Integration: Incorporates custom tools for tasks such as data visualization and external data retrieval, enhancing the agents' capabilities.
Ensure you have Python 3.8+ installed on your system. Some knowledge of LangChain, CrewAI, and machine learning concepts is beneficial but not strictly necessary.
- Clone the repository:
git clone https://github.com/vlcekl/ds-crew.git - Navigate to the project directory:
cd project-name - Install the required dependencies:
pip install -r requirements.txt
Refer to the notebooks/quickstart_guide.ipynb for a step-by-step introduction to setting up and running a basic agent crew.
- Defining Agents: Create new agents by extending the base Agent class in the
agents/directory, specifying their roles and capabilities. - Creating Crews: Combine agents into crews within the
crew/directory, defining the collaborative tasks and objectives. - Running the System: Execute the main crew script to start the system:
python crew/main_crew.py
Feel free to contribute to the project by submitting pull requests or opening issues for bugs, suggestions, or enhancements.
Run the tests to ensure the system's integrity:
pytest tests/
For more detailed information on LangChain and CrewAI, visit the official documentation:
- LangChain: LangChain Documentation
- CrewAI: CrewAI Guide
This project is licensed under the MIT License - see the LICENSE file for details.
- Thanks to the LangChain and CrewAI teams for providing the frameworks that made this project possible.
- Special thanks to all contributors who have helped shape this project.