ViperEkura 8150ab6c32 feat: add torch.compile CLI option for training ...

- Add --compile flag (default/reduce-overhead/max-autotune)
- Apply torch.compile in _before_wrap before DDP/FSDP wrapping
- Profiling shows MFU 85.5% -> 88.5% (+3%), time -3.2%, memory -7.9%

2026-07-29 22:06:51 +08:00

..

__init__.py

chore : 升级 torch 2.11.0+cu128，移除自定义 Muon，修复 gloo device_id

2026-06-27 16:10:37 +08:00

metric_util.py

refactor : grad_norm 指标简化，clip_grad_norm 移至 executor

2026-06-30 14:59:43 +08:00

rollout.py

fix: correct online rollout lifecycle

2026-07-23 19:01:37 +08:00

schedule.py

fix: scheduler division-by-zero, loss_mask bool

2026-07-03 22:04:55 +08:00

strategy.py

chore: fix ruff lint warnings and signal handling edge cases

2026-07-25 21:08:30 +08:00

train_callback.py

fix: keep metric logs cumulative instead of segmental in each checkpoint

2026-07-28 09:18:48 +08:00

train_context.py

feat: add torch.compile CLI option for training

2026-07-29 22:06:51 +08:00

trainer.py

refactor: move signal_handler from parallel/ to top-level for broader reuse

2026-07-28 10:36:17 +08:00