Phitter AI Agent

Phitter is a web application and Python library developed by our team. It is designed to identify the most suitable probability distribution for any given dataset. The tool offers an extensive range of features, including over 80 different probability distributions, goodness-of-fit tests, and interactive visualizations, making it an invaluable resource for statisticians, data scientists, and researchers who need to analyze and model data accurately. In our current project, we are extending the capabilities of Phitter by developing an intelligent agent that can interact with the application using natural language. This approach aims to make Phitter more accessible and user-friendly, allowing users to leverage its powerful functionalities without needing to write complex code or have deep statistical knowledge. To achieve this, we are utilizing LLaMA3, a state-of-the-art language model, and agents from LLaMA Index, a framework that enhances the interaction between language models and specific applications. The integration of these technologies will enable users to input their queries and receive detailed, context-aware responses from Phitter, streamlining the process of data analysis and distribution fitting. URL Website: https://phitter.io/ Repository: https://github.com/phitterio/phitter-kernel Python Package: https://pypi.org/project/phitter/