Cantitate/Preț
Produs

Introducing MLOps

Autor Mark Treveil, Nicolas Omont, Clément Stenac, Kenji Lefevre, Du Phan, Joachim Zentici, Adrien Lavoillotte, Makoto Miyazaki, Lynn Heidmann
en Limba Engleză Paperback – 5 ian 2021

Notăm cu interes faptul că Introducing MLOps abordează direct marea problemă a departamentelor de date moderne: faptul că majoritatea modelelor de învățare automată rămân la stadiul de prototip. Metodologia propusă de cei nouă autori de la O'Reilly se concentrează pe arhitectură și guvernanță, transformând un proces adesea haotic într-o disciplină inginerească riguroasă. Structura cărții urmărește cele cinci etape critice ale ciclului de viață ML — de la construcție și preproducție până la monitorizare și mentenanță pe termen lung. Găsim în această carte un echilibru necesar între soluțiile tehnice și strategiile organizaționale. Autorii, printre care se numără Mark Treveil și Nicolas Omont, explică modul în care procesele de MLOps pot reduce riscurile prin modele explicabile și corecte, asigurând în același timp o automatizare fluidă a pipeline-urilor. Abordarea diferă de Practical MLOps de Noah Gift prin faptul că este mai puțin concentrată pe un set specific de tool-uri cloud (cum ar fi AWS sau Azure) și mai mult axată pe principiile aplicabile universal pentru operaționalizarea modelelor în sisteme de business complexe. Credem că valoarea adăugată a acestui volum constă în perspectiva sa pragmatică asupra mentenanței. Nu este suficient să lansezi un model; acesta trebuie recalibrat și monitorizat constant pentru a-și păstra acuratețea. Tonul tehnic și orientat spre specificații concrete face din această lucrare un ghid esențial pentru echipele care doresc să treacă de la experimente izolate la un impact real în afaceri, oferind soluții pentru eliminarea barierelor dintre data scientists și inginerii de aplicații.

Citește tot Restrânge

Preț: 30350 lei

Preț vechi: 37936 lei
-20%

Puncte Express: 455

Carte disponibilă

Livrare economică 26 septembrie-10 octombrie

Livrare prin curier în România Termenul estimat este afișat lângă disponibilitate.
Transport gratuit de la 40000 lei Plată online sau ramburs, în funcție de opțiunile comenzii.
Retur gratuit în 14 zile Comandă securizată și suport în română.

Specificații

ISBN-13: 9781492083290
ISBN-10: 1492083291
Pagini: 183
Dimensiuni: 177 x 228 x 12 mm
Greutate: 0.35 kg
Ediția:1
Editura: O'Reilly

De ce să citești această carte

Recomandăm această carte inginerilor de machine learning și specialiștilor în data science care vor să asigure succesul modelelor lor în producție. Veți câștiga o metodologie clară pentru a gestiona întregul ciclu de viață ML, reducând riscurile organizaționale și tehnice. Este un ghid practic pentru oricine dorește să transforme algoritmii complecși în instrumente de business fiabile și scalabile.


Descriere

More than half of the analytics and machine learning (ML) models created by organizations today never make it into production. Some of the challenges and barriers to operationalization are technical, but others are organizational. Either way, the bottom line is that models not in production can't provide business impact. This book introduces the key concepts of MLOps to help data scientists and application engineers not only operationalize ML models to drive real business change but also maintain and improve those models over time. Through lessons based on numerous MLOps applications around the world, nine experts in machine learning provide insights into the five steps of the model life cycle--Build, Preproduction, Deployment, Monitoring, and Governance--uncovering how robust MLOps processes can be infused throughout. This book helps you: Fulfill data science value by reducing friction throughout ML pipelines and workflows Refine ML models through retraining, periodic tuning, and complete remodeling to ensure long-term accuracy Design the MLOps life cycle to minimize organizational risks with models that are unbiased, fair, and explainable Operationalize ML models for pipeline deployment and for external business systems that are more complex and less standardized

Notă biografică

Mark Treveil has designed products in fields as diverse as telecoms, banking, and online trading. His own startup led a revolution in governance in the UK local government, where it still dominates. He is now part of the Dataiku Product Team based in Paris.
Nicolas Omont is VP of operations at Artelys where he is developing mathematical optimization solutions for energy and transport. He previously held the role of Dataiku Product Manager for ML and advanced analytics. He holds a PhD in Computer Science, and he's been working in operations research and statistics for the past 15 years, mainly in the telecommunications and energy utility sectors.
Clément Stenac is a passionate software engineer, CTO and co-founder at Dataiku. He oversees the design, development of the Dataiku DSS Entreprise AI Platform. Clément was previously head of product development at Exalead, leading the design and implementation of web-scale search engine software. He also has extensive experience with open source software, as a former developer of the VideoLAN (VLC) and Debian projects.
Kenji Lefevre is VP Product at Dataiku. He oversees the product roadmap and the user experience of the Dataiku DSS Entreprise AI Platform. He holds a PhD in pure mathematics from University of Paris VII, and he directed documentary movies before switching to Data Science and product management.
Du Phan is a Machine Learning engineer at Dataiku, where he works in democratizing data science. In the past few years, he has been dealing with a variety of data problems, from geospatial analysis to deep learning. His work now focuses on different facets and challenges of MLOps.
Joachim Zentici is an Engineering Director at Dataiku. Joachim graduated in applied mathematics from Ecole Centrale Paris. Prior to joining Dataiku in 2014, he was a Research Engineer in computer vision at Siemens Molecular Imaging and INRIA. He has also been a teacher and a lecturer. At Dataiku, Joachim had multiple contributions including managing the engineers in charge of the core infrastructure, building the team for the plugins & ecosystem effort as well as leading the global technology training program for customer-facing engineers.
Adrien Lavoillotte is Engineering Director at Dataiku where he leads the team responsible for machine learning and statistics features in the software. He studied at ECE Paris, a graduate school of engineering, and worked for several startups before joining Dataiku in 2015.
Makoto Miyazaki is a Data Scientist at Dataiku and responsible for delivering hands-on consulting services using Dataiku DSS for European and Japanese clients. Makoto holds a Bachelor's degree in economics and a Master's Degree in data science, and he was also a former financial journalist with a wide range of beats, including nuclear energy and economic recoveries from the tsunami.
Lynn Heidmann received her Bachelor of Arts in Journalism/Mass Communications and Anthropology from the University of Wisconsin-Madison in 2008 and decided to bring her passion for research and writing into the world of tech. She spent seven years in the San Francisco Bay Area writing and running operations with Google and subsequently Niantic before moving to Paris to head content initiatives at Dataiku. In her current role, Lynn follows and writes about technological trends and developments in the world of data and AI.