Code

AI Training - Max#59

Code

Merged

profvjreddi merged 51 commits intoharvard-edge:mainharvard-edge/cs249r_book:mainfrom agnusmaximus:trainingagnusmaximus/cs249r_book:trainingCopy head branch name to clipboardDec 7, 2023

+3,660-3,325

Commits

Commits on Nov 14, 2023

• First pass at training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  3e8ced1

• Upd

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  5c26613

Commits on Nov 15, 2023

• Added section place holders

  profvjreddi

  committed
  5af7468

• Updated the introduction

  profvjreddi

  committed
  624e6c1

• Add overview

  profvjreddi

  committed
  bf96aca

• Updated introduction

  profvjreddi

  committed
  ac1f659

• updated NN intro

  profvjreddi

  committed
  aab7018

• Minor rework of the writing

  profvjreddi

  committed
  fb27e38

• Merge branch 'main' into pr/59

  profvjreddi

  committed
  5c0b895

• Added training data content

  profvjreddi

  committed
  65b4812

• Focusing on the big items we cover in the chapter

  profvjreddi

  committed
  3c67582

Commits on Nov 20, 2023

• Update backprop

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  d3902ee

• Markdown

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  307dafb

• Upd training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  cf3b67f

• Upd training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  0d0c613

• Upd training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  893b750

• Upd training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  fe2a424

• Upd training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  79757d1

• Upd training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  733ec36

Commits on Nov 27, 2023

• Merge branch 'main' into pr/59

  profvjreddi

  committed
  c58a6d0

• aitraining images

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  c7be42a

Commits on Nov 30, 2023

• Expanded the algorithms section and added references

  profvjreddi

  committed
  20fe3b7

• Merge branch 'main' into training

  profvjreddi

  authored
  752d3e4

• Updated cover image

  profvjreddi

  committed
  e382921

• Update training.qmd

  profvjreddi

  committed
  199779e

• Remove grammar fix

  profvjreddi

  committed
  a38cddd

• Made a pass on the hyperparameter section

  profvjreddi

  committed
  7412fe1

Commits on Dec 5, 2023

• Merge branch 'main' into pr/59

  profvjreddi

  committed
  7beb911

• References training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  c2e48ec

• Adding in more details about regularization

  profvjreddi

  committed
  bdc4f3f

• Minor updates to Regularization section

  profvjreddi

  committed
  55e4171

• Weight initialization update pass

  profvjreddi

  committed
  49794e9

• Making weight initialization connections to weight init

  profvjreddi

  committed
  076a599

• Update training.qmd

  profvjreddi

  committed
  4f7c6bc

• fix bibtex reference for he et al.

  profvjreddi

  committed
  3fe19f4

• Added details to activation functions

  profvjreddi

  committed
  bfbdc4c

• references

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  7307e84

Commits on Dec 6, 2023

• references bib

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  f5c468a

• references

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  ea66794

• Improving the MM section with details

  profvjreddi

  committed
  cecadb0

• MD fixes

  profvjreddi

  committed
  14f6950

• Merge branch 'training' of https://github.com/agnusmaximus/cs249r_book into pr/59

  profvjreddi

  committed
  ad05af0

• sorting references

  profvjreddi

  committed
  a60904a

• Improving the optimizations section

  profvjreddi

  committed
  1bc8513

Commits on Dec 7, 2023

• Roofline / big batch-size training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  a42d79d

• training

  

  Maximilian Lam

  authored and

  Maximilian Lam

  committed
  d92a50b

• Updated the training parallelization section

  profvjreddi

  committed
  a16a060

• Merge branch 'training' of https://github.com/agnusmaximus/cs249r_book into pr/59-training

  profvjreddi

  committed
  bff3e9e

• Updated the learning objectives.

  profvjreddi

  committed
  606cee9

• Updated the conclusion section.

  profvjreddi

  committed
  8c0c3bf

• MD fixes

  profvjreddi

  committed
  5817210