Code

Ptc refactor#65

Code

Merged

cssprad1 merged 42 commits intodevelopnasa-nccs-hpda/pytorch-caney:developfrom ptc-refactornasa-nccs-hpda/pytorch-caney:ptc-refactorCopy head branch name to clipboardNov 8, 2024

+2,100-7,970

Commits

Commits on Apr 4, 2024

• added updated frontier code

  

  Caleb Spradlin

  committed
  bcc151b

• rmd tar pieces

  

  Caleb Spradlin

  committed
  8ccdd90

Commits on Jun 7, 2024

• updated pre-training code

  

  Caleb S. Spradlin

  committed
  7221ae0

• added frontier code

  

  Caleb Spradlin

  committed
  b84c587

Commits on Jun 10, 2024

• added latest code, runners, and notebooks

  

  cssprad1

  committed
  4f1f2d2

• added note

  

  Caleb Spradlin

  committed
  79548c5

Commits on Jul 3, 2024

• per step checkpointing

  

  cssprad1

  committed
  3b3d10c

• fixed bug with checkpointing

  

  cssprad1

  committed
  8c99c72

• updated data path

  

  cssprad1

  committed
  422b7bd

Commits on Aug 26, 2024

• added config changes for hackathon

  

  Caleb Spradlin

  committed
  30a5734

• made more config changes

  

  Caleb Spradlin

  committed
  f433e11

Commits on Aug 27, 2024

• changed batch size for example script

  

  cssprad1

  committed
  7e72fd7

• added tensorboard notebook

  

  cssprad1

  committed
  5b435b6

• add tensorboard

  

  Jian Li

  committed
  add1cb1

• Merge branch 'ht24_jl' into hackathon_2024

  Show description for 45ee11e
  

  Jian Li

  committed
  45ee11e

• Merge branch 'hackathon_2024' of https://github.com/nasa-nccs-hpda/pytorch-caney into hackathon_2024

  

  Jian Li

  committed
  84ac000

Commits on Aug 28, 2024

• added tensorboard updates

  

  cssprad1

  committed
  aa846b0

• made tensorboard dir arg

  

  cssprad1

  committed
  80ec8a2

• made config changes

  

  Caleb Spradlin

  committed
  11342e2

• Adding optimizer module and lamb with fusedlamb with deepspeed

  

  Jordan Caraballo-Vega

  committed
  2de0b35

• Merge pull request #57 from nasa-nccs-hpda/hackaton_2024_jacaraba

  Show description for 265f252
  jordancaraballo

  authored
  265f252

• merged jordans changes

  

  cssprad1

  committed
  5c6ad7c

• added lamb gradient logging

  

  cssprad1

  committed
  d01a940

• added lamb gradient logging

  

  cssprad1

  committed
  2211008

• rmd address for slurm

  

  cssprad1

  committed
  f0597d7

• Merge pull request #58 from nasa-nccs-hpda/hackathon_2024_cssprad1

  Show description for 089c81c
  cssprad1

  authored
  089c81c

Commits on Sep 6, 2024

• add DeepSpeed Flops Profiler to tensorboard

  

  Jian Li

  committed
  2dbbc9b

• add DeepSpeed Flops Profiler to tensorboard

  

  Jian Li

  committed
  b370537

• new TAG

  

  Jian Li

  committed
  a9df1a6

Commits on Sep 9, 2024

• Merge pull request #60 from nasa-nccs-hpda/hackaton_2024_jl

  Show description for 341b414
  cssprad1

  authored
  341b414

Commits on Sep 16, 2024

• test updates from matched pair code assesment

  

  cssprad1

  committed
  5a7c5d5

• Adding code coverage through match pair trial

  jordancaraballo

  committed
  a6cc223

• Merge remote-tracking branch 'caleb/matched-pair-cssprad1' into hackathon_2024

  jordancaraballo

  committed
  d66072b

• Some README documentation for regression test locally

  jordancaraballo

  committed
  75ca214

Commits on Oct 22, 2024

• removed files, changed structure, initial commit

  

  cssprad1

  committed
  a851f86

• init files for all subdirs

  

  cssprad1

  committed
  d1591c9

Commits on Oct 23, 2024

• initial pretraining commit

  

  cssprad1

  committed
  fa0aa69

Commits on Oct 30, 2024

• added model handling and 3dcloud task

  

  cssprad1

  committed
  7e6430e

Commits on Nov 6, 2024

• updated pipelines

  

  cssprad1

  committed
  8f8d9b1

• deleted other examples

  

  cssprad1

  committed
  72f0197

Commits on Nov 8, 2024

• merge conflicts

  

  cssprad1

  committed
  5dc47bf

• merge conflicts

  

  cssprad1

  committed
  9a7cd6f