PyAutoFit: A Classy Probabilistic Programming Language for Model Composition and Fitting

TL;DR

PyAutoFit is a comprehensive Python-based probabilistic programming language that simplifies model specification, fitting, and analysis, supporting complex model composition, customization, and large-scale result management.

Contribution

It introduces PyAutoFit, a versatile PPL that integrates model building, fitting, visualization, and result analysis within a unified framework, including advanced features like database tools and domain-specific knowledge integration.

Findings

Supports high-dimensional model composition

Enables customizable fitting procedures

Includes tools for large-scale result analysis

Abstract

A major trend in academia and data science is the rapid adoption of Bayesian statistics for data analysis and modeling, leading to the development of probabilistic programming languages (PPL). A PPL provides a framework that allows users to easily specify a probabilistic model and perform inference automatically. PyAutoFit is a Python-based PPL which interfaces with all aspects of the modeling (e.g., the model, data, fitting procedure, visualization, results) and therefore provides complete management of every aspect of modeling. This includes composing high-dimensionality models from individual model components, customizing the fitting procedure and performing data augmentation before a model-fit. Advanced features include database tools for analysing large suites of modeling results and exploiting domain-specific knowledge of a problem via non-linear search chaining. Accompanying PyAutoFit is the autofit workspace (see https://github.com/Jammy2211/autofit_workspace), which includes example scripts and the HowToFit lecture series which introduces non-experts to model-fitting and provides a guide on how to begin a project using PyAutoFit. Readers can try PyAutoFit right now by going to the introduction Jupyter notebook on Binder (see https://mybinder.org/v2/gh/Jammy2211/autofit_workspace/HEAD) or checkout our readthedocs(see https://pyautofit.readthedocs.io/en/latest/) for a complete overview of PyAutoFit's features.