Filters








71 Hits in 4.8 sec

On Continuous Integration / Continuous Delivery for Automated Deployment of Machine Learning Models using MLOps [article]

Satvik Garg, Pradyumn Pundir, Geetanjali Rathee, P.K. Gupta, Somya Garg, Saransh Ahlawat
2022 arXiv   pre-print
Model deployment in machine learning has emerged as an intriguing field of research in recent years. It is comparable to the procedure defined for conventional software development.  ...  In the MLOps approach, we discuss tools and approaches for executing the CI/CD pipeline of machine learning frameworks.  ...  MLOps Level 1 The steps of machine learning experiments are orchestrated at MLOps level 1, to automate the ML pipeline solely to undertake continuous model training.  ... 
arXiv:2202.03541v1 fatcat:c4r2vpisjvb4haqgvbkmw4inlm

Machine Learning Operations: A Survey on MLOps Tool Support [article]

Nipuni Hewage, Dulani Meedeniya
2022 arXiv   pre-print
Therefore, the importance of the Machine Learning Operations (MLOps) concept, which can deliver appropriate solutions for such concerns, is discussed.  ...  Finally, we recognize that there is a shortage in the availability of a fully functional MLOps platform on which processes can be automated by reducing human intervention.  ...  The integration of machine learning (ML) practices that support data engineering, with the DevOps based software development, has resulted in Machine Learning Operations (MLOps).  ... 
arXiv:2202.10169v1 fatcat:alo4yg2w5vh2dbfwvncxi3klmu

Applying DevOps Practices of Continuous Automation for Machine Learning

Ioannis Karamitsos, Saeed Albarhami, Charalampos Apostolopoulos
2020 Information  
The machine learning processes of development and deployment during the experimentation phase may seem easy.  ...  debt, improve value delivery and maintenance, and improve operational functions for real-world machine learning applications.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/info11070363 fatcat:kqhl6bfm7vhsbi4yu4c7xidtqi

Application of DevOps in the improvement of machine learning processes

Beatriz Mayumi Andrade Matsui, Denise Hideko Goya
2020 Zenodo  
DevOps practices have been increasingly used by software development teams to automate and simplify processes ranging from integration, through testing, approval, implementation, to the final delivery  ...  The present study aims to focus on the possibility of applying this concept also in teams that work with machine learning and could benefit from the improvements brought with the adoption of DevOps.  ...  of machine learning models.  ... 
doi:10.5281/zenodo.4318113 fatcat:si6hb7zb6veg5nfqde62iajkru

MLOps: A Taxonomy and a Methodology

Matteo Testi, Matteo Ballabio, Emanuele Frontoni, Giulio Iannello, Sara Moccia, Paolo Soda, Gennaro Vessio
2022 IEEE Access  
The pipeline is based on ten steps: business problem understanding, data acquisition, ML methodology, ML training & testing, continuous integration, continuous delivery, continuous training, continuous  ...  ML Operations (MLOps) represents an effective strategy for bringing ML models from academic resources to useful tools for solving problems in the corporate world.  ...  Continuous integration for ML systems relies on having a substantial impact on the end-to-end pipeline to automate the delivery of the ML models with minimal effort.  ... 
doi:10.1109/access.2022.3181730 fatcat:4gokhtootvacjc2xjgeqkhrwhq

Machine Learning Operations (MLOps): Overview, Definition, and Architecture [article]

Dominik Kreuzberger, Niklas Kühl, Sebastian Hirschl
2022 arXiv   pre-print
The paradigm of Machine Learning Operations (MLOps) addresses this issue. MLOps includes several aspects, such as best practices, sets of concepts, and development culture.  ...  The final goal of all industrial machine learning (ML) projects is to develop ML products and rapidly bring them into production.  ...  It ensures automation with continuous integration, continuous delivery, and continuous deployment (CI/CD), thus allowing for fast, frequent, and reliable releases.  ... 
arXiv:2205.02302v3 fatcat:bzermjilzvh35bpmqnv2yfzvh4

Demystifying MLOps and Presenting a Recipe for the Selection of Open-Source Tools

Philipp Ruf, Manav Madan, Christoph Reich, Djaffar Ould-Abdeslam
2021 Applied Sciences  
Deep learning has revolutionized the field of Image processing, and building an automated machine learning workflow for object detection is of great interest for many organizations.  ...  Nowadays, machine learning projects have become more and more relevant to various real-world use cases.  ...  Acknowledgments: The contents of this publication are taken from the research project "(Q-AMeLiA)-Quality Assurance of Machine Learning Applications", which is supervised by Hochschule Furtwangen University  ... 
doi:10.3390/app11198861 fatcat:phrlxsiieve2vcltz5i2v5lamm

Reliable Fleet Analytics for Edge IoT Solutions [article]

Emmanuel Raj, Magnus Westerlund, Leonardo Espinosa-Leal
2021 arXiv   pre-print
In this paper, we propose a framework for facilitating machine learning at the edge for AIoT applications, to enable continuous delivery, deployment, and monitoring of machine learning models at the edge  ...  For the machine learning experiments, we forecast multivariate time series for predicting air quality in the respective rooms by using the models deployed in respective edge devices.  ...  Automation for machine learning based systems is driven by seamless monitoring, continuous integration and continuous delivery as following: a) Continuous Integration (CI) and Continuous Delivery (CD):  ... 
arXiv:2101.04414v1 fatcat:2v6fe2ugbjbilcyiymy32kiziy

Scaling Enterprise Recommender Systems for Decentralization [article]

Maurits van der Goes
2021 arXiv   pre-print
Creating a culture of self-service, automation, and collaboration to scale recommender systems for decentralization.  ...  We demonstrate a practical use case of a self-service ML workspace deployment and a recommender system, that scale faster to subsidiaries and with less technical debt.  ...  For example by releasing solutions with a continuous integration and continuous deployment pipeline. • Data availability: Access to validated datasets via a feature store and indexed in a data catalog  ... 
arXiv:2109.09231v1 fatcat:lg6op2hs3jgk3mvawam7kwaniy

Who Needs MLOps: What Data Scientists Seek to Accomplish and How Can MLOps Help? [article]

Sasu Mäkinen and Henrik Skogström and Eero Laaksonen and Tommi Mikkonen
2021 arXiv   pre-print
Following continuous software engineering practices, there has been an increasing interest in rapid deployment of machine learning (ML) features, called MLOps.  ...  In the results, the majority of respondents are in category (i) or (ii), focusing on data and models; however the benefits of MLOps only emerge in category (iii) when there is a need for frequent retraining  ...  ACKNOWLEDGEMENT The authors would sincerely like to thank Valohai (https://valohai.com/) for an access to survey data, and Business Finland (project AIGA -AI Governance and Auditing) for supporting this  ... 
arXiv:2103.08942v1 fatcat:jn66ocw2ynfstfl3hdt45qnxiq

Who Needs MLOps: What Data Scientists Seek to Accomplish and How Can MLOps Help?

Sasu Makinen, Henrik Skogstrom, Eero Laaksonen, Tommi Mikkonen
2021 2021 IEEE/ACM 1st Workshop on AI Engineering - Software Engineering for AI (WAIN)  
Following continuous software engineering practices, there has been an increasing interest in rapid deployment of machine learning (ML) features, called MLOps.  ...  In the results, the majority of respondents are in category (i) or (ii), focusing on data and models; however the benefits of MLOps only emerge in category (iii) when there is a need for frequent retraining  ...  ACKNOWLEDGEMENT The authors would sincerely like to thank Valohai (https://valohai.com/) for an access to survey data, and Business Finland (project AIGA -AI Governance and Auditing) for supporting this  ... 
doi:10.1109/wain52551.2021.00024 fatcat:jtqf57zllvgjbo3e7vo3qrolua

Towards Regulatory-Compliant MLOps: Orazivio's Journey from a Machine Learning Experiment to a Deployed Certified Medical Product

Tuomas Granlund, Vlad Stirbu, Tommi Mikkonen
2021 SN Computer Science  
Examples of such features include machine learning (ML) models, which are usually pre-trained, but can still evolve in production.  ...  In this paper, we start by presenting continuous software engineering practices in a regulated context, and then apply the results to the emerging practice of MLOps, or continuous delivery of ML features  ...  Acknowledgements The authors would like to thank Business Finland and the members of the AHMED (Agile and Holistic MEdical software Development) consortium for their contribution in preparing this paper  ... 
doi:10.1007/s42979-021-00726-1 fatcat:tefiffu7gve7tm5v56c24ezahm

Continual-Learning-as-a-Service (CLaaS): On-Demand Efficient Adaptation of Predictive Models [article]

Rudy Semola, Vincenzo Lomonaco, Davide Bacciu
2022 arXiv   pre-print
Specifically, it embraces continual machine learning and continuous integration techniques.  ...  It provides support for model updating and validation tools for data scientists without an on-premise solution and in an efficient, stateful and easy-to-use manner.  ...  A core feature not present in machine learning tools for the MLOps process is the use of the Continual Learning metric to detect model of the performance decay over time.  ... 
arXiv:2206.06957v2 fatcat:dbs7fmql6ffjppept27knysii4

MLOps Challenges in Multi-Organization Setup: Experiences from Two Real-World Cases

Tuomas Granlund, Aleksi Kopponen, Vlad Stirbu, Lalli Myllyaho, Tommi Mikkonen
2021 2021 IEEE/ACM 1st Workshop on AI Engineering - Software Engineering for AI (WAIN)  
The setup we assume is that of continuous deployment, often referred to DevOps in software development. When also ML components are deployed in a similar fashion, term MLOps is used.  ...  Since this data can be confidential in nature, it cannot always be openly shared in seek of artificial intelligence (AI) and machine learning (ML) solutions.  ...  ACKNOWLEDGEMENT The authors would like to thank Business Finland and the members of AHMED (Agile and Holistic MEdical software Development) & AIGA (AI Governance and Auditing) consortiums for their contribution  ... 
doi:10.1109/wain52551.2021.00019 fatcat:6thcxcery5dxfiblot3q7t2fza

Agility in Software 2.0 – Notebook Interfaces and MLOps with Buttresses and Rebars [article]

Markus Borg
2021 arXiv   pre-print
Due to the experimental approach used by data scientists when developing machine learning models, agility is an essential characteristic.  ...  Solutions based on machine learning bring both great opportunities, thus coined "Software 2.0," but also great challenges for the engineering community to tackle.  ...  Our thanks go to Backtick Technologies for hosting the MSc thesis project and Dr. Niklas Fors, Dept. of Computer Science, Lund University for acting as the examiner.  ... 
arXiv:2111.14142v1 fatcat:alm4wx5dtfhldldurpl5hz26gi
« Previous Showing results 1 — 15 out of 71 results