幫助中心 | 我的帳號 | 關於我們

R語言簡潔建模(影印版)(英文版)

  • 作者:(美)馬克斯·庫恩//朱莉婭·斯萊格|責編:張燁
  • 出版社:東南大學
  • ISBN:9787576605907
  • 出版日期:2023/03/01
  • 裝幀:平裝
  • 頁數:363
人民幣:RMB 118 元      售價:
放入購物車
加入收藏夾

內容大鋼
    tidymodels是一組用於建模和機器學習的R語言包。無論你是新手還是擁有多年建模經驗,這本實踐用書將為數據分析師、業務分析師和數據科學家展示tidymodels框架如何為你的工作提供一致、靈活的方法。
    RStudio的工程師Max kuhn和Julia Silge展示了通過專註于一種被稱為tidyverse的R方言來創建模型的方法。採用tidyverse原則的軟體共享高層設計理念和低層語法及數據結構,因此學習生態系統的一部分有助於掌握下一部分。你會明白為什麼tidymodels框架被人們廣泛使用。
    本書內容包括:
    ·學習從頭到尾構建模型所需的步驟
    ·理解如何流暢地使用不同的建模和特徵工程方法
    ·研究如何避免建模的常見缺陷,比如過擬合
    ·學習為建模準備數據的實用方法
    ·調整模型以獲得最佳性能
    ·使用良好的統計實踐來比較、評估和選擇模型

作者介紹
(美)馬克斯·庫恩//朱莉婭·斯萊格|責編:張燁

目錄
Preface
Part I.  Introduction
  1. Software for Modeling
    Fundamentals for Modeling Software
    Types of Models
      Descriptive Models
      Inferential Models
      Predictive Models
    Connections Between Types of Models
    Some Terminology
    How Does Modeling Fit into the Data Analysis Process?
    Chapter Summary
  2. A Tiflyverse Primer
    Tidyverse Principles
      Design for Humans
      Reuse Existing Data Structures
      Design for the Pipe and Functional Programming
    Examples of Tidyverse Syntax
    Chapter Summary
  3. A Review of R Modeling Fundamentals
    An Example
    What Does the R Formula Do?
    Why Tidiness Is Important for Modeling
    Combining Base R Models and the Tidyverse
    The tidymodels Metapackage
    Chapter Summary
Part II.  Modeling Basics
  4. The Ames Housing Data
    Exploring Features of Homes in Ames
    Chapter Summary
  5. Spending Our Data
    Common Methods for Splitting Data
    What About a Validation Set?
    Multilevel Data
    Other Considerations for a Data Budget
    Chapter Summary
  6. Fitting Models with parsnip
    Create a Model
    Use the Model Results
    Make Predictions
    parsnip-Extension Packages
    Creating Model Specifications
    Chapter Summary
  7. A Model Workflow
    Where Does the Model Begin and End?
    Workflow Basics
    Adding Raw Variables to the workflow0
    How Does a workflow0 Use the Formula?
      Tree-Based Models
      Special Formulas and Inline Functions

    Creating Multiple Workflows at Once
    Evaluating the Test Set
    Chapter Summary
  8. Feature Engineering with Recipes
    A Simple recipe() for the Ames Housing Data
    Using Recipes
    How Data Are Used by the recipe()
    Examples of Steps
      Encoding Qualitative Data in a Numeric Format
      Interaction Terms
      Spline Functions
      Feature Extraction
      Row Sampling Steps
      General Transformations
      Natural Language Processing
    Skipping Steps for New Data
    Tidy a recipe()
    Column Roles
    Chapter Summary
  9. Judging Model Effectiveness
    Performance Metrics and Inference
    Regression Metrics
    Binary Classification Metrics
    Multiclass Classification Metrics
    Chapter Summary
Part Ill.  Tools for Creating Effective Models
  10. Resampling for Evaluating Performance
    The Resubstitution Approach
    Resampling Methods
      Cross-Validation
      Repeated Cross-Validation
      Leave-One-Out Cross-Validation
      Monte Carlo Cross-Validation
      Validation Sets
      Bootstrapping
      Rolling Forecasting Origin Resampling
    Estimating Performance
    Parallel Processing
    Saving the Resampled Objects
    Chapter Summary
  11. Comparing Models with Resampling
    Creating Multiple Models with Workflow Sets
    Comparing Resampled Performance Statistics
    Simple Hypothesis Testing Methods
    Bayesian Methods
      A Random Intercept Model
      The Effect of the Amount of Resampling
    Chapter Summary
  12. Model Tuning and the Dangers of Overntting
    Model Parameters

    Tuning Parameters for Different Types of Models
    What Do We Optimize?
    The Consequences of Poor Parameter Estimates
    Two General Strategies for Optimization
    Tuning Parameters in tidymodels
    Chapter Summary
  13. Grid Search
    Regular and Nonregular Grids
      Regular Grids
      Nonregular Grids
    Evaluating the Grid
    Finalizing the Model
    Tools for Creating Tuning Specifications
    Tools for Efficient Grid Search
      Submodel Optimization
      Parallel Processing
      Benchmarking Boosted Trees
      Access to Global Variables
      Racing Methods
    Chapter Summary
  14. Iterative Search
    A Support Vector Machine Model
    Bayesian Optimization
      A Gaussian Process Model
      Acquisition Functions
      The tune_bayes() Function
    Simulated Annealing
      Simulated Annealing Search Process
      The tune_sim_anneal() Function
    Chapter Summary
  15. Screening Many Models
    Modeling Concrete Mixture Strength
    Creating the Workflow Set
    Tuning and Evaluating the Models
    Efficiently Screening Models
    Finalizing a Model
    Chapter Summary
Part IV.  Beyond the Basics
  16. Dimensionality Reduction
    What Problems Can Dimensionality Reduction Solve?
    A Picture Is Worth a Thousand...Beans
    A Starter Recipe
    Recipes in the Wild
      Preparing a Recipe
      Baking the Recipe
    Feature Extraction Techniques
      Principal Component Analysis
      Partial Least Squares
      Independent Component Analysis
      Uniform Manifold Approximation and Projection

    Modeling
    Chapter Summary
  17. Encoding Categorical Data
    Is an Encoding Necessary?
    Encoding Ordinal Predictors
    Using the Outcome for Encoding Predictors
      Effect Encodings in tidymodels
      Effect Encodings with Partial Pooling
    Feature Hashing
    More Encoding Options
    Chapter Summary
  18. Explaining Models and Predictions
    Software for Model Explanations
    Local Explanations
    Global Explanations
    Building Global Explanations from Local Explanations
    Back to Beans!
    Chapter Summary
  19. When Should You Trust Your Predictions?
    Equivocal Results
    Determining Model Applicability
    Chapter Summary
  20. Ensembles of Models
    Creating the Training Set for Stacking
    Blend the Predictions
    Fit the Member Models
    Test Set Results
    Chapter Summary
  21. Inferential Analysis
    Inference for Count Data
    Comparisons with Two-Sample Tests
    Log-Linear Models
    A More Complex Model
    More Inferential Analysis
    Chapter Summary
Appendix. Recommended Preprocessing
References
Index

  • 商品搜索:
  • | 高級搜索
首頁新手上路客服中心關於我們聯絡我們Top↑
Copyrightc 1999~2008 美商天龍國際圖書股份有限公司 臺灣分公司. All rights reserved.
營業地址:臺北市中正區重慶南路一段103號1F 105號1F-2F
讀者服務部電話:02-2381-2033 02-2381-1863 時間:週一-週五 10:00-17:00
 服務信箱:bookuu@69book.com 客戶、意見信箱:cs@69book.com
ICP證:浙B2-20060032