Kdump 24afceddb7 [trainer] fix wsd scheduler (#7304) ...

* [trainer] Warmup_stable_decay supports setting the number of stable and decay steps according to the warmup_ratio ratio

* Update trainer_utils.py

---------

Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>

2025-03-26 15:25:02 +08:00

..

dpo

[trainer] fix vlm loss for transformers 4.49 (#7448)

2025-03-24 10:24:05 +08:00

kto

[trainer] fix vlm loss for transformers 4.49 (#7448)

2025-03-24 10:24:05 +08:00

ppo

[misc] update format (#7277)

2025-03-13 02:53:08 +08:00

pt

[trainer] fix vlm loss for transformers 4.49 (#7448)

2025-03-24 10:24:05 +08:00

rm

[misc] update format (#7277)

2025-03-13 02:53:08 +08:00

sft

[trainer] fix vlm loss for transformers 4.49 (#7448)

2025-03-24 10:24:05 +08:00

__init__.py

rename package

2024-05-16 18:39:08 +08:00

callbacks.py

[model] add qwen2vl 32b & upgrade peft (#7469)

2025-03-25 12:15:58 +08:00

test_utils.py

[misc] update format (#7277)

2025-03-13 02:53:08 +08:00

trainer_utils.py

[trainer] fix wsd scheduler (#7304)

2025-03-26 15:25:02 +08:00

tuner.py

[3rdparty] fix redundant process group destroy for ray (#7395)

2025-03-21 10:56:47 +08:00