Kubeflow in Action
Manning Publications (Verlag)
978-1-61729-913-1 (ISBN)
In Kubeflow in Action you will learn how to:
Set up interactive data science notebooks in a modern containerized environment
Create pipelines for complex data processing and model training jobs
Serve models in production for applications
Train models on large datasets using distributed computing resources
Use distributed resources to optimize model hyperparameters and architectures
Make production deployments
Kubeflow in Action is an authoritative hands-on guide to deploying machine learning to production using the Kubeflow MLOps platform. Each stage of the ML workflow is explored and illustrated with engaging use cases that are based on tasks regularly tackled by data scientists. You’ll learn how Kubeflow can support training models, automatically tune hyperparameters, and speed up reinforcement learning. Each use case is explored from both a data science and software engineering perspective, so you benefit from a 360° understanding of Kubeflow. about the technology Kubeflow is a Kubernetes-native MLOps platform that easily installs on clusters and is instantly familiar to anyone who’s worked with Kubernetes. It’s perfect for handling large-scale infrastructure for AI workloads, with support for distributed machine learning training and complex data pipelines. about the book Kubeflow in Action shows you how to utilize Kubeflow to rapidly scale machine learning projects from a laptop to a distributed cluster. You’ll kick off with a rapid introduction to containers, benefit from careful guidance on Kubeflow’s installation and initial setup, and master core Kubeflow tasks like storing data, training models, and monitoring metrics.
Detailed use cases help show how to construct complex pipelines, automate hyperparameter tuning, and implement network architecture search. You’ll quickly progress to a deep dive into Kubeflow’s more advanced uses, including training distributed models, deployment, A/B testing, and infrastructure monitoring to help trigger actions based on incoming data. about the reader For data scientists and data engineers. Data scientists will benefit from learning to handle data pipeline infrastructure issues as well as scaling techniques. Data engineers will learn to streamline their daily tasks and support their scientist colleagues.
Juana Nakfour is a principal software engineer at Red Hat, in the AI Services product team. She is also the technical and community lead for Open Data Hub, an open-source project providing an end-to-end AI/ML platform including Kubeflow on Openshift. Sanjay Arora is a senior principal software engineer and data scientist in the AI Center of Excellence at Red Hat. In the past, Sanjay has worked at CERN, Goldman Sachs, and Palantir.
table of contents READ IN LIVEBOOK 1DATA SCIENCE AT SCALE READ IN LIVEBOOK 2GETTING STARTED READ IN LIVEBOOK 3KUBEFLOW, AN END-TO-END AI/ML PLATFORM READ IN LIVEBOOK 4PIPELINES A DEEPER LOOK READ IN LIVEBOOK 5DISTRIBUTED MODEL TRAINING: INTRODUCTION READ IN LIVEBOOK 6DISTRIBUTED MODEL TRAINING: ADVANCED 7 DEPLOYING AND MONITORING MODELS EPILOGUE APPENDIXES READ IN LIVEBOOK APPENDIX A: INSTALLING KUBEFLOW
| Erscheinungsdatum | 17.05.2023 |
|---|---|
| Verlagsort | New York |
| Sprache | englisch |
| Themenwelt | Mathematik / Informatik ► Informatik ► Datenbanken |
| Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik | |
| ISBN-10 | 1-61729-913-8 / 1617299138 |
| ISBN-13 | 978-1-61729-913-1 / 9781617299131 |
| Zustand | Neuware |
| Informationen gemäß Produktsicherheitsverordnung (GPSR) | |
| Haben Sie eine Frage zum Produkt? |
aus dem Bereich
