Bücher von Valliappa Lakshmanan

The design patterns in this book capture best practices and solutions to recurring problems in machine learning. The authors, three Google engineers, catalog proven methods to help data scientists tackle common problems throughout the ML process. These design patterns codify the experience of hundreds of experts into straightforward, approachable advice.In this book, you will find detailed explanations of 30 patterns for data and problem representation, operationalization, repeatability, reproducibility, flexibility, explainability, and fairness. Each pattern includes a description of the problem, a variety of potential solutions, and recommendations for choosing the best technique for your situation.You'll learn how to:Identify and mitigate common challenges when training, evaluating, and deploying ML modelsRepresent data for different ML model types, including embeddings, feature crosses, and moreChoose the right model type for specific problemsBuild a robust training loop that uses checkpoints, distribution strategy, and hyperparameter tuningDeploy scalable ML systems that you can retrain and update to reflect new dataInterpret model predictions for stakeholders and ensure models are treating users fairly

This practical book shows you how to employ machine learning models to extract information from images. ML engineers and data scientists will learn how to solve a variety of image problems including classification, object detection, autoencoders, image generation, counting, and captioning with proven ML techniques. This book provides a great introduction to end-to-end deep learning: dataset creation, data preprocessing, model design, model training, evaluation, deployment, and interpretability.Google engineers Valliappa Lakshmanan, Martin Grner, and Ryan Gillard show you how to develop accurate and explainable computer vision ML models and put them into large-scale production using robust ML architecture in a flexible and maintainable way. You'll learn how to design, train, evaluate, and predict with models written in TensorFlow or Keras.You'll learn how to:Design ML architecture for computer vision tasksSelect a model (such as ResNet, SqueezeNet, or EfficientNet) appropriate to your taskCreate an end-to-end ML pipeline to train, evaluate, deploy, and explain your modelPreprocess images for data augmentation and to support learnabilityIncorporate explainability and responsible AI best practicesDeploy image models as web services or on edge devicesMonitor and manage ML models

Bewährte Praxislösungen für komplexe Machine-Learning-Aufgaben Behandelt alle Phasen der ML-Produktpipeline Klar strukturierter Aufbau, der dafür sorgt, dass sich Konzepte und Zusammenhänge rasch erschließen Fokus auf TensorFlow, aber auch übertragbar auf PyTorch-Projekte Die Design Patterns in diesem Buch zeigen praxiserprobte Lösungen für wiederkehrende Aufgaben im Machine Learning. Die Autor:innen - ML-Experten bei Google - beschreiben Methoden, die Data Scientists helfen, typische Probleme im gesamten ML-Prozess zu bewältigen. Die Entwurfsmuster verdichten die Erfahrungen von Hunderten von Expert:innen zu klar strukturierten, zugänglichen Best Practices. Das Buch bietet detaillierte Erläuterungen zu 30 Mustern für die Daten- und Problemdarstellung, Operationalisierung, Wiederholbarkeit, Reproduzierbarkeit, Flexibilität, Erklärbarkeit und Fairness. Zu jedem Muster erhalten Sie eine Beschreibung des Problems, eine Vielzahl möglicher Lösungen sowie Empfehlungen, welche Technik die beste für Ihre Problemstellung ist. Erfahren Sie, wie Sie: Herausforderungen beim Trainieren, Bewerten und Deployen von ML-Modellen erkennen und überwinden Daten für verschiedene ML-Modelltypen mit Einbettungen, Feature Crosses und mehr darstellen den richtigen Modelltyp für bestimmte Fragestellungen auswählen eine robuste Trainingsschleife mit Checkpoints, Verteilungsstrategie und Hyperparameter-Tuning erstellen skalierbare ML-Systeme deployen, die bei erneutem Training aktuelle Daten berücksichtigen Modellvorhersagen für Stakeholder interpretieren Modellgenauigkeit, Reproduzierbarkeit, Resilienz und Fairness verbessern

The ability to create automated algorithms to process gridded spatial data is increasingly important as remotely sensed datasets increase in volume and frequency.

The ability to create automated algorithms to process gridded spatial data is increasingly important as remotely sensed datasets increase in volume and frequency.