ViperEkura e8ff7f5321 fix: use batch_per_device for rollout scheduler batch sizing ...

- train_context.py referenced non-existent cfg.batch_size, replaced with cfg.batch_per_device
- default group_size lowered from 8 to 1: without a group concept (DPO), scheduler batch equals batch_per_device; rollout-based DPO can opt in via extra_kwargs['group_size']>=2
- inline expressions (rollout_batch_size, max_seq_len) extracted for readability
- add tests/trainer/test_online_e2e.py: end-to-end online_dpo via Trainer.train, exercising KV-cache-backed rollout path

2026-07-20 13:32:04 +08:00

..

conftest.py

refactor : replace iteration with consumed_samples

2026-06-30 18:42:42 +08:00

test_callbacks.py

refactor : 清理工厂和配置系统中的死代码与冗余抽象

2026-06-07 11:39:50 +08:00

test_early_stopping.py

feat: add --resume flag to decouple weight loading from training resumption

2026-07-16 14:23:23 +08:00

test_grpo_strategy.py

refactor: separate old policy and ref model in GRPO strategy

2026-07-14 20:03:45 +08:00

test_online_e2e.py

fix: use batch_per_device for rollout scheduler batch sizing

2026-07-20 13:32:04 +08:00

test_online_strategy.py

refactor: unify rollout onto inference engine KV-cache path

2026-07-20 12:52:20 +08:00

test_rollout.py

refactor: unify rollout onto inference engine KV-cache path

2026-07-20 12:52:20 +08:00

test_train_strategy.py

refactor : 清理工厂和配置系统中的死代码与冗余抽象

2026-06-07 11:39:50 +08:00

test_trainer.py

refactor: 检查点加载重构，路径替代对象传递

2026-05-27 20:15:29 +08:00