Commit Graph

9 Commits

Author SHA1 Message Date
晓雷 44947fe7fa ray launcher before test 2025-07-15 14:22:45 +08:00
bowei.fw 888751da38 qwen2 grpo works 2025-07-12 09:44:50 +08:00
bowei.fw 0cbddb8aba . 2025-07-11 11:10:55 +08:00
bowei.fw a78fd2dd24 rm mb_spec 2025-07-09 22:16:31 +08:00
bowei.fw 32077b02ed add close 2025-07-09 21:56:27 +08:00
bowei.fw b75aaaf130 refactor trainer 2025-07-09 21:48:28 +08:00
bowei.fw 124184a7b7 minor fix 2025-07-09 16:48:11 +08:00
bowei.fw a6bcab22ba support fsdp engine and sglang remote engine 2025-07-09 16:45:29 +08:00
博惟 15dfbe837c PullRequest: 332 [lite] Support FSDP engines
Merge branch mzy/lite/engines of git@code.alipay.com:inclusionAI/AReaL.git into lite
https://code.alipay.com/inclusionAI/AReaL/pull_requests/332

Reviewed-by: 晓雷 <meizhiyu.mzy@antgroup.com>


* fsdp2 engine
* fix utils
* add fsdp engine test
* .
* fsdp engine test passed
* unsqueeze immediately before model inputs and after model outputts
* add optimizer save/load, add position id calculation for input
* .
* format
* not to squeeze
* add train and eval api
* .
* .
* improve fsdp engine data preprocessing
* format
* PullRequest: 337 [lite] Add SFT trainer example.
* trainer log
* minor changes
* add update weights from disk
* fix type annotation
2025-07-09 16:24:25 +08:00