mirror of https://github.com/hpcaitech/ColossalAI.git synced 2026-07-16 17:16:14 +00:00

Files

YeAnbang 14f237ce7e [feat] Support boxed math reward (#6284 )

* fix pp+tp, fix dataloader

* fixed plugin micro-batch size

* support boxed reward

* add boxed reward

* fix pp state dict incomplete issue

* Revert "fix pp state dict incomplete issue"

This reverts commit 6c1b3b694f.

2025-04-29 16:46:47 +08:00

reward

[feat] Support boxed math reward (#6284 )

2025-04-29 16:46:47 +08:00

__init__.py

[chat] add distributed impl (#6210 )

2025-02-21 15:24:23 +08:00

comm.py

[chat] add distributed impl (#6210 )

2025-02-21 15:24:23 +08:00

consumer.py

[feat] Support boxed math reward (#6284 )

2025-04-29 16:46:47 +08:00

grpo_consumer.py

[feat] Support boxed math reward (#6284 )

2025-04-29 16:46:47 +08:00

inference_backend.py

[feat] Support DAPO (#6263 )

2025-04-25 17:39:17 +08:00

launch.py

fix pp+tp, fix dataloader (#6280 )

2025-04-28 17:10:00 +08:00

loss.py

[feat] Support DAPO (#6263 )

2025-04-25 17:39:17 +08:00

producer.py

fix pp+tp, fix dataloader (#6280 )

2025-04-28 17:10:00 +08:00

README.md

[chat] add distributed impl (#6210 )

2025-02-21 15:24:23 +08:00

utils.py

[feat] Support DAPO (#6263 )

2025-04-25 17:39:17 +08:00

README.md

Requirements

pip install cupy-cuda12x
python -m cupyx.tools.install_library --cuda 12.x --library nccl