ViperEkura 853aaeefde fix: always leave a checkpoint when training is interrupted ...

- save an emergency checkpoint even when the signal is handled before the first optimizer step, where optimizer_step == last_ckpt_step used to skip the save entirely (CI race on slow cold-start runners)
- track a saved-this-session flag so interrupted runs always have at least one checkpoint, while normal zero-step completion keeps skipping the save

2026-09-03 20:36:50 +08:00

..

config

fix: resolve audited dispatch, kernel, and rollout bugs

2026-09-03 16:58:14 +08:00

dataset

refactor: deduplicate low-risk code paths

2026-08-19 16:17:40 +08:00

extension

fix: resolve audited dispatch, kernel, and rollout bugs

2026-09-03 16:58:14 +08:00

inference

fix: resolve audited training, import, and serving bugs

2026-09-03 20:27:41 +08:00

model

fix: resolve audited training, import, and serving bugs

2026-09-03 20:27:41 +08:00

optim

fix: resolve audited training, import, and serving bugs

2026-09-03 20:27:41 +08:00

parallel

fix: resolve audited training, import, and serving bugs

2026-09-03 20:27:41 +08:00

preprocessing

fix: resolve audited training, import, and serving bugs

2026-09-03 20:27:41 +08:00

serialization

fix: resolve audited training, import, and serving bugs

2026-09-03 20:27:41 +08:00

tokenize

refactor: simplify BaseFactory and separate ModelFactory from AutoModel

2026-07-30 09:38:20 +08:00

trainer

fix: always leave a checkpoint when training is interrupted

2026-09-03 20:36:50 +08:00

__init__.py

refactor: simplify training and inference interfaces

2026-08-19 20:55:13 +08:00

factory.py

refactor: simplify BaseFactory and separate ModelFactory from AutoModel

2026-07-30 09:38:20 +08:00

logging.py

fix: report true rank in logs via dist-aware helpers

2026-08-27 13:39:38 +08:00

protocols.py

refactor: 重构训练后端为 Executor 模式

2026-05-24 20:35:44 +08:00

signal_handler.py

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

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