PyMilo: A Python Library for ML I/O

TL;DR

PyMilo is a Python library that offers a transparent, safe, and reliable way to serialize, export, and deploy machine learning models, addressing safety and transparency issues of existing formats.

Contribution

PyMilo introduces a non-executable, transparent serialization format for ML models, improving safety and ease of deployment over traditional binary formats.

Findings

Provides a reliable model export/import solution

Enhances safety and transparency in model deployment

Simplifies ML model development pipeline

Abstract

PyMilo is an open-source Python package that addresses the limitations of existing Machine Learning (ML) model storage formats by providing a transparent, reliable, and safe method for exporting and deploying trained models. Current formats, such as pickle and other binary formats, have significant problems, such as reliability, safety, and transparency issues. In contrast, PyMilo serializes ML models in a transparent non-executable format, enabling straightforward and safe model exchange, while also facilitating the deserialization and deployment of exported models in production environments. This package aims to provide a seamless, end-to-end solution for the exportation and importation of pre-trained ML models, which simplifies the model development and deployment pipeline.