设计机器学习系统(影印版)
Chip Huyen
出版时间:2022年09月
页数:367
“简而言之,这是一本关于如何在公司构建、部署和扩展机器学习模型以获得最大影响的佳作。”
一Josh Wills
WeaveGrid软件工程师,Slack前数据工程总监
“在一个蓬勃发展却又混乱的生态系统中,这种关于端到端机器学习的原则性观点既是你的地图,也是你的指南针:无论你是否身处大型科技公司,这都是一本必读之作。”
--Jacopo Tagliabue
Coveo人工智能总监

机器学习系统既复杂又独特。复杂是因为包含大量组件,涉及许多不同的利益方;独特是因为其依赖于数据,不同用例之间的数据差异很大。在本书中,你将学习以一种整体方法来设计兼具可靠性、可伸缩性、可维护性,并能适应不断变化的环境和业务需求的机器学习系统。
作者Chip Huyen是Claypot Al的联合创始人,她在如何帮助系统作为一个整体实现其目标的背景下考虑了每一种设计决策,例如如何处理和创建训练数据,使用哪些特性,重新训练模型的频率,以及监测哪些内容。书中的迭代框架采用了真实的案例研究,并辅以大量参考资料。
这本书将帮助你处理以下情况:
● 工程化数据并选择正确的指标来解决业务问题
● 实现持续开发、评估、部署和更新模型的流程自动化
● 开发监控系统,快速检测和解决模型在生产中可能遇到的问题
● 构建跨用例服务的机器学习平台
● 开发可靠的机器学习系统
  1. Preface
  2. 1. Overview of Machine Learning Systems
  3. When to Use Machine Learning
  4. Understanding Machine Learning Systems
  5. Summary
  6. 2. Introduction to Machine Learning Systems Design
  7. Business and ML Objectives
  8. Requirements for ML Systems
  9. Iterative Process
  10. Framing ML Problems
  11. Mind Versus Data
  12. Summary
  13. 3. Data Engineering Fundamentals
  14. Data Sources
  15. Data Formats
  16. Data Models
  17. Data Storage Engines and Processing
  18. Modes of Dataflow
  19. Batch Processing Versus Stream Processing
  20. Summary
  21. 4. Training Data
  22. Sampling
  23. Labeling
  24. Class Imbalance
  25. Data Augmentation
  26. Summary
  27. 5. Feature Engineering
  28. Learned Features Versus Engineered Features
  29. Common Feature Engineering Operations
  30. Data Leakage
  31. Engineering Good Features
  32. Summary
  33. 6. Model Development and Offline Evaluation
  34. Model Development and Training
  35. Model Offline Evaluation
  36. Summary
  37. 7. Model Deployment and Prediction Service
  38. Machine Learning Deployment Myths
  39. Batch Prediction Versus Online Prediction
  40. Model Compression
  41. ML on the Cloud and on the Edge
  42. Summary
  43. 8. Data Distribution Shifts and Monitoring
  44. Causes of ML System Failures
  45. Data Distribution Shifts
  46. Monitoring and Observability
  47. Summary
  48. 9. Continual Learning and Test in Production
  49. Continual Learning
  50. Test in Production
  51. Summary
  52. 10. Infrastructure and Tooling for MLOps
  53. Storage and Compute
  54. Development Environment
  55. Resource Management
  56. ML Platform
  57. Build Versus Buy
  58. Summary
  59. 11. The Human Side of Machine Learning
  60. User Experience
  61. Team Structure
  62. Responsible AI
  63. Summary
  64. Epilogue
  65. Index
书名:设计机器学习系统(影印版)
作者:Chip Huyen
国内出版社:东南大学出版社
出版时间:2022年09月
页数:367
书号:978-7-5766-0224-1
原版书书名:Designing Machine Learning Systems
原版书出版商:O'Reilly Media
Chip Huyen
 
Chip Huyen是实时机器学习平台Claypot AI的联合创始人。凭借在 NVIDIA、Netflix和Snorkel Al的工作,她帮助了一些世界上最大的组织开发和部署机器学习系统。本书是Chip根据她在斯坦福大学开设的课程“机器学习系统设计”(CS329S)的讲义撰写的。
 
 
The animal on the cover of Designing Machine Learning Systems is a red-legged partridge (Alectoris rufa), also known as a French partridge.
Bred for centuries as a gamebird, this economically important, largely nonmigratory member of the pheasant family is native to western continental Europe, though populations have been introduced elsewhere, including England, Ireland, and New Zealand.
Relatively small but stout bodied, the red-legged partridge boasts ornate coloration and feather patterning, with light brown to gray plumage along its back, a light pink belly, a cream-colored throat, a brilliant red bill, and rufous or black barring on its flanks.
Feeding primarily on seeds, leaves, grasses, and roots, but also on insects, red-legged partridges breed each year in dry lowland areas, such as farmland, laying their eggs in ground nests. Though they continue to be bred in large numbers, these birds are now considered near threatened due to steep population declines attributed, in part, to overhunting and disappearance of habitat. Like all animals on O’Reilly covers, they’re vitally important to our world.
购买选项
定价:138.00元
书号:978-7-5766-0224-1
出版社:东南大学出版社