whylogs: Embrace Data Logging Across Your ML Systems

whylogs is an open-source data logging library that enables data logging for any ML/AI pipeline in a few lines of code. It provides essential insights into the health and performance of the pipeline, helping developers and data scientists to be aware of changes in the quality and consistency of their data. The library supports a wide range of ML data types, including complex inputs like images and video, and is designed for both interactive analysis and production settings. whylogs is massively scalable due to its static memory footprint and lightweight output, which minimizes storage costs. It also prevents overcollection of data to reduce costs, security vulnerabilities, and operational complexity. The library currently supports columnar data in Python and Java, with plans to add support for time series, text, and image data types, as well as more languages.