Active Machine Learning with Python : Refine and Elevate Data Quality over Quantity with Active Learning

Active Machine Learning with Python Refine and Elevate Data Quality over Quantity with Active Learning

Use active machine learning with Python to improve the accuracy of predictive models, streamline the data analysis process, and adapt to evolving data trends, fostering innovation and progress across diverse fields Key Features Learn how to implement a pipeline for optimal model creation from large...

Descripción completa

Detalles Bibliográficos
Otros Autores: Masson-Forsythe, Margaux, author (author)
Formato: Libro electrónico
Idioma:Inglés
Publicado: Birmingham, England : Packt Publishing [2024]
Edición:First edition
Materias:
Machine learning.
Python (Computer program language)
Data mining.
Ver en Biblioteca Universitat Ramon Llull:https://discovery.url.edu/permalink/34CSUC_URL/1im36ta/alma991009810644406719
Tabla de Contenidos:
  • Cover
  • Title Page
  • Copyright and Credits
  • Contributors
  • Table of Contents
  • Preface
  • Part 1: Fundamentals of Active Machine Learning
  • Chapter 1: Introducing Active Machine Learning
  • Understanding active machine learning systems
  • Definition
  • Potential range of applications
  • Key components of active machine learning systems
  • Exploring query strategies scenarios
  • Membership query synthesis
  • Stream-based selective sampling
  • Pool-based sampling
  • Comparing active and passive learning
  • Summary
  • Chapter 2: Designing Query Strategy Frameworks
  • Technical requirements
  • Exploring uncertainty sampling methods
  • Understanding query-by-committee approaches
  • Maximum disagreement
  • Vote entropy
  • Average KL divergence
  • Labeling with EMC sampling
  • Sampling with EER
  • Understanding density-weighted sampling methods
  • Summary
  • Chapter 3: Managing the Human in the Loop
  • Technical requirements
  • Designing interactive learning systems and workflows
  • Exploring human-in-the-loop labeling tools
  • Common labeling platforms
  • Handling model-label disagreements
  • Programmatically identifying mismatches
  • Manual review of conflicts
  • Effectively managing human-in-the-loop systems
  • Ensuring annotation quality and dataset balance
  • Assess annotator skills
  • Use multiple annotators
  • Balanced sampling
  • Summary
  • Part 2: Active Machine Learning in Practice
  • Chapter 4: Applying Active Learning to Computer Vision
  • Technical requirements
  • Implementing active ML for an image classification project
  • Building a CNN for the CIFAR dataset
  • Applying uncertainty sampling to improve classification performance
  • Applying active ML to an object detection project
  • Preparing and training our model
  • Analyzing the evaluation metrics
  • Implementing an active ML strategy.
  • Using active ML for a segmentation project
  • Summary
  • Chapter 5: Leveraging Active Learning for Big Data
  • Technical requirements
  • Implementing ML models for video analysis
  • Selecting the most informative frames with Lightly
  • Using Lightly to select the best frames to label for object detection
  • SSL with active ML
  • Summary
  • Part 3: Applying Active Machine Learning to Real-World Projects
  • Chapter 6: Evaluating and Enhancing Efficiency
  • Technical requirements
  • Creating efficient active ML pipelines
  • Monitoring active ML pipelines
  • Determining when to stop active ML runs
  • Enhancing production model monitoring with active ML
  • Challenges in monitoring production models
  • Active ML to monitor models in production
  • Early detection for data drift and model decay
  • Summary
  • Chapter 7: Utilizing Tools and Packages for Active ML
  • Technical requirements
  • Mastering Python packages for enhanced active ML
  • scikit-learn
  • modAL
  • Getting familiar with the active ML tools
  • Summary
  • Index
  • Other Books You May Enjoy.

Ejemplares similares