Part 1: Core PyTorch

This book offers a practical, start-to-finish guide to building deep learning projects using the PyTorch library. Readers will learn fundamental concepts such as tensors and neural network architectures, gain hands-on experience in training models from examples, and discover how to deploy solutions for real-world applications. Through major projects like developing language models, image generation systems, and medical image segmentation, the content equips learners with essential skills for a career in artificial intelligence development.

This page is free — overview and chapter list only.
The complete book body is sold separately.

Full book: $2.99 USDC via x402 ·
Per chapter: $0.25 USDC

Buy / open complete book (HTML)
· Complete book Markdown (.md)

Agents: start with free /library/discovery.json, sample free teaser chapters,
then pay for individual chapters or the complete book URL above.
Append .md to any content URL for Markdown with YAML front matter.

Chapters

  1. About The Cover Illustration (Free teaser)
  2. 16 Hardware And Software Requirements ($0.25)
  3. 21 A Pretrained Network That Recognizes The Subject Of An Im ($0.25)
  4. 22 Generating And Editing Images ($0.25)
  5. 24 A Pretrained Network That Describes Scenes ($0.25)
  6. Summary ($0.25)
  7. 32 Tensors Multidimensional Arrays ($0.25)
  8. 35 Named Tensors ($0.25)
  9. 36 Tensor Element Types ($0.25)
  10. 391 Views Of Another Tensors Storage ($0.25)
  11. 3101 Managing A Tensors Device Attribute ($0.25)
  12. This Chapter Covers ($0.25)
  13. 413 Changing The Layout ($0.25)
  14. 433 Representing Scores ($0.25)
  15. 443 Ready For Training ($0.25)
  16. 455 Text Embeddings As A Blueprint ($0.25)
  17. 524 Choosing A Linear Model As A First Try ($0.25)
  18. Broadcasting ($0.25)
  19. 543 Iterating To Fit The Model ($0.25)
  20. 544 Normalizing Inputs ($0.25)
  21. 551 Computing The Gradient Automatically ($0.25)
  22. 552 Optimizers La Carte ($0.25)
  23. 555 Autograd Nits And Switching It Off ($0.25)
  24. 613 Adding Nonlinearity With Activation Functions ($0.25)
  25. 616 What Learning Means For A Neural Network ($0.25)
  26. 622 Returning To The Linear Model ($0.25)
  27. 633 Comparing To The Linear Model ($0.25)
  28. 711 Downloading Cifar 10 ($0.25)
  29. 714 Normalizing Data ($0.25)
  30. 724 Representing The Output As Probabilities ($0.25)
  31. 725 Training The Classifier ($0.25)
  32. 73 Conclusion ($0.25)
  33. 82 Convolutions In Action ($0.25)
  34. 823 Looking Further With Depth And Pooling ($0.25)
  35. 83 Subclassing Nn Module ($0.25)
  36. 84 Training Our Convolutional Neural Network ($0.25)
  37. 85 Model Design ($0.25)
  38. Batch Normalization ($0.25)
  39. 91 A Motivating Example Generating Names Character By Charac ($0.25)
  40. 93 Generating Our Training Data ($0.25)
  41. 95 Attention ($0.25)
  42. 952 Scaled Dot Product Causal Self Attention ($0.25)
  43. Gpt Style Decoder ($0.25)
  44. Transformer Encoder ($0.25)
  45. 99 The Vision Transformer ($0.25)
  46. Summary 2 ($0.25)
  47. 105 The Forward Process ($0.25)
  48. 107 Reversing Diffusion How To Sample ($0.25)
  49. 108 Conclusion ($0.25)
  50. 113 What Is A Ct Scan Exactly ($0.25)
  51. On The Shoulders Of Giants ($0.25)
  52. 1142 Our Data Source The Luna Grand Challenge ($0.25)
  53. 122 Parsing Lunas Annotation Data ($0.25)
  54. 123 Loading Individual Ct Scans ($0.25)
  55. 1244 Extracting A Nodule From A Ct Scan ($0.25)
  56. 1254 Rendering The Data ($0.25)
  57. Dataparallel Vs Distributeddataparallel ($0.25)
  58. 1341 The Core Convolutions ($0.25)
  59. 1342 The Full Model ($0.25)
  60. 1351 The Computebatchloss Function ($0.25)
  61. Listing 1317 Masking For Classification Metrics ($0.25)
  62. 1372 Interlude The Tqdm Function ($0.25)
  63. 1462 Seeing The Improvement From Data Augmentation ($0.25)
  64. 151 Utilizing A Second Model In Our Project ($0.25)
  65. 1531 The Segment Anything Model Sam ($0.25)
  66. 1541 Trying Out An Off The Shelf Model For Our Project ($0.25)
  67. 1563 Training A Model To Flag Potential Candidates ($0.25)
  68. 1571 How To Fine Tune A Model ($0.25)
  69. 1574 Saving Our Model ($0.25)
  70. 1613 Initializing A Distributed Program ($0.25)
  71. Listing 169 Creating A Data Loader With Distributedsampler ($0.25)
  72. 1651 Pipeline Parallelism ($0.25)
  73. Listing 1614 Synchronizing Model Parameters ($0.25)
  74. 1682 Expert Parallelism ($0.25)
  75. This Chapter Covers 2 ($0.25)
  76. 1712 Our Model Behind A Fastapi Server ($0.25)
  77. Implementation ($0.25)
  78. Add Quantization ($0.25)
  79. 1721 Interoperability Beyond Pytorch With Onnx ($0.25)
  80. History And Motivations Behind Torchcompile ($0.25)