Commits · 3fe2519852b9a7f43fba33ec03c82337abc2fc6a · Simulation and Data Lab Applied Machine Learning / PyTorch at JSC

Browse files

Mar 06, 2025
- Clean up whitespace · 3fe25198
  Jan Ebert authored 5 months ago
  
  3fe25198
Feb 20, 2025
- Fix deprecation warnings · 6303cd82
  Jose Ignacio Robledo authored 6 months ago and Jan Ebert committed 5 months ago
  6303cd82
Nov 14, 2024
- Add code for later PyTorch versions · ceb61d4f
  Jan Ebert authored 9 months ago
  ceb61d4f
- Include singular loading function · b4c87a7f
  Jan Ebert authored 9 months ago
  b4c87a7f
Nov 03, 2024
- Fix train step logging · 871386bd
  Jan Ebert authored 9 months ago
  
  871386bd
- Do not shuffle test/validation splits · 20148fbf
  Jan Ebert authored 9 months ago
  20148fbf
Oct 25, 2024
- Rename distributed checkpointing import · 68e35585
  Jan Ebert authored 9 months ago
  68e35585
Oct 16, 2024
- Specify Gloo backend for CPU · bd050edf
  Jan Ebert authored 10 months ago
  bd050edf
- Properly shut down distributed process group · 1721a1aa
  Jan Ebert authored 10 months ago
  1721a1aa
Oct 09, 2024
- Also scale other optimizer parameters · c941cc9c
  Jan Ebert authored 10 months ago
  c941cc9c
Jul 09, 2024
- Use square root LR scaling · e72c8976
  Jan Ebert authored Jul 9, 2024
  e72c8976
- Use a simple auto wrapping policy · 1707daa1
  Jan Ebert authored Jul 9, 2024
  1707daa1
Jul 02, 2024
- Extend `no_grad` context · f1c468df
  Jan Ebert authored Jul 2, 2024
  
  f1c468df
- Make learning rate scaling optional non-default · cdbbbbf1
  Jan Ebert authored Jul 2, 2024
  cdbbbbf1
May 03, 2024
- Add FSDP example · 1a3c1694
  Jan Ebert authored May 3, 2024
  1a3c1694
- Add PyTorch DDP example · eac57cd5
  Jan Ebert authored May 3, 2024
  
  eac57cd5