What question did this study set out to answer?

The research aims to develop an adaptable framework for monitoring machine learning processes efficiently.

March 10, 2026Open Access

ObServML: Deployable Python application for compact and modular systems monitoring

Key Points

The research aims to develop an adaptable framework for monitoring machine learning processes efficiently.
Developed a microservices-based framework for monitoring.
Integrated with RabbitMQ for real-time data communication.
Utilized MLflow for model versioning and management.
Implemented various machine learning techniques such as decision trees and autoencoders.
Enabled effective fault detection and isolation in various operations.
Facilitated real-time monitoring through robust data communication.
Supported adaptive and modular monitoring solutions, enhancing predictive maintenance capabilities.

Abstract

ObservML enables the combination of training and deploying ML monitoring models within a single microservices-based system. Its application focuses on monitoring problems that can be solved with fault detection and isolation (FDI), time series analysis, and process mining through an operator-friendly and adaptable framework based on MLOps practices. The framework is developed to connect to RabbitMQ for real-time data communication and MLflow for model versioning. It supports a wide range of machine learning techniques, including decision trees, autoencoders, and time series models, providing a robust toolkit for anomaly detection and predictive maintenance, and can be extended as required.

KI fragen

Bookmark

View Full Paper