YeAnbang
7ee4452f8c
fix vllm
2025-03-19 17:11:10 +08:00
duanjunwen
7795d4c50d
[Feature] Support Distributed LogProb for GRPO Training ( #6247 )
...
* [fix] fix qwen VocabParallelLMHead1D and gather output
* fix tp bug
* fix consumer
* [feat] Support Distributed LogProb for GRPO Training
* [fix] fix loss func
* [fix] fix log prob plugin
* [fix] fix qwen modeling param
* [fix] rm comments
* [fix] rm hard-code;fix non-dist version
* [fix] fix test file param name and benchmark tp gather output=True/False
* [fix] rm non-dist version in dist log prob
* [fix] fix comments
* [fix] fix dis log prob plugin
* [fix] fix test case
* [fix] fix qwen VocabParallelLMHead1D and gather output
* [fix] fix DistLogProb comments
* [fix] restore tp size
* [fix] fix comments
* [fix] fix comment; fix LogSoftmax usage
---------
Co-authored-by: Tong Li <tong.li35271158@gmail.com>
2025-03-18 17:47:55 +08:00
YeAnbang
bc0171d392
fix transformers backend
2025-03-14 18:12:35 +08:00
Tong Li
57b49da5e4
setup update
2025-03-13 16:52:15 +08:00
Tong Li
45ac6c6cb2
print results
2025-03-13 16:51:22 +08:00
Tong Li
4702d57841
convert to 8 generation
2025-03-13 16:49:02 +08:00
Tong Li
afddfde2dd
fix consumer
2025-03-13 14:55:26 +08:00
Tong Li
131eeceb5d
fix tp bug
2025-03-13 14:52:09 +08:00
Tong Li
704866a240
detach
2025-03-11 16:17:02 +08:00
Tong Li
47d6493778
add response length
2025-03-11 13:06:09 +08:00
Tong Li
abca66e69f
fix reward score
2025-03-11 10:17:32 +08:00
Tong Li
71a0181fce
update reward
2025-03-10 14:19:10 +08:00
Tong Li
754b16dfbf
update reward fn
2025-03-10 14:18:22 +08:00
Tong Li
9d9d51614e
update grpo
2025-03-10 14:12:04 +08:00
pre-commit-ci[bot]
eb6337f07f
[pre-commit.ci] auto fixes from pre-commit.com hooks
...
for more information, see https://pre-commit.ci
2025-03-06 08:29:59 +00:00
Tong Li
22cc1558a8
Merge branch 'grpo-latest' of github.com:hpcaitech/ColossalAI into grpo-latest
2025-03-06 16:28:47 +08:00
Tong Li
0590f10fb7
update select algo
2025-03-06 16:27:13 +08:00
Tong Li
0cc0c843ed
add save
2025-03-06 16:26:14 +08:00
pre-commit-ci[bot]
ab5b6d8432
[pre-commit.ci] auto fixes from pre-commit.com hooks
...
for more information, see https://pre-commit.ci
2025-03-06 06:30:27 +00:00
Tong Li
0f566cc2d4
add algo selection
2025-03-06 14:29:22 +08:00
Tong Li
812f4b7750
update loader
2025-03-06 11:44:42 +08:00
Tong Li
7f2ceac5c3
update example
2025-03-06 10:54:23 +08:00
Tong Li
d03cdea949
update reward fn
2025-03-06 10:53:48 +08:00
Tong Li
678f5a9eca
update loss
2025-03-06 10:53:03 +08:00
Tong Li
b96d69055e
grpo consumer
2025-03-06 10:51:27 +08:00
Tong Li
c15225bc52
modify data loader
2025-03-06 10:49:44 +08:00
Tong Li
070907dd7f
polish
2025-02-28 10:16:42 +08:00
Tong Li
f736d747e3
update grpo
2025-02-25 18:12:04 +08:00
Tong Li
ffd3878a1e
add simple grpo
2025-02-23 22:54:26 +08:00
Tong Li
8e6c9a4ab3
add reward related function
2025-02-23 11:02:54 +08:00
Hongxin Liu
de282dd694
[feature] fit RL style generation ( #6213 )
...
* [feature] fit rl style generation
* [doc] add docstr
* [doc] add docstr
2025-02-21 17:28:19 +08:00
Hongxin Liu
43c9b5fb44
[chat] add distributed impl ( #6210 )
2025-02-21 15:24:23 +08:00
Hongxin Liu
9379cbd668
[release] update version ( #6195 )
...
* [release] update version
* fix test
* fix test
2025-02-20 11:36:18 +08:00
binmakeswell
24dee8f0b7
[doc] DeepSeek V3/R1 news ( #6199 )
...
* [doc] DeepSeek V3/R1 news
* [doc] DeepSeek V3/R1 news
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-02-19 15:07:29 +08:00
Hongxin Liu
f73ae55394
[application] add lora sft example data ( #6198 )
2025-02-18 20:18:18 +08:00
Tong Li
f8b9e88484
[application] Update README ( #6196 )
...
* remove unused ray
* remove unused readme
* update readme
* update readme
* update
* update
* add link
* update readme
* update readme
* fix link
* update code
* update cititaion
* update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update readme
* update project
* add images
* update link
* update note
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-02-18 20:17:56 +08:00
Hongxin Liu
d54642a263
[application] add lora sft example ( #6192 )
...
* [application] add lora sft example
* update requirements
* update readme
* update comment
* update ci
2025-02-18 13:06:38 +08:00
YeAnbang
d20c8ffd97
Add GRPO and Support RLVR for PPO ( #6186 )
...
* add grpo, support rlvr
* add grpo, support rlvr
* tested deepseek r1 pipeline
* add ci
* verify grpo r1
* verify grpo r1
* update readme, remove unused code
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove path
* clean code
* fix circular import
* fix ci OOM
* fix ci OOM
* skip kto tp, fix qwen generation
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-02-18 09:43:36 +08:00
flybird11111
ce0ec40811
[checkpointio] fix for async io ( #6189 )
2025-02-14 17:34:13 +08:00
Hongxin Liu
5ff5323538
[hotfix] fix zero optim save ( #6191 )
2025-02-14 15:09:50 +08:00
Hongxin Liu
014837e725
[shardformer] support pipeline for deepseek v3 and optimize lora save ( #6188 )
...
* [shardformer] support pipeline for deepseek v3
* [checkpointio] fix lora save
* [devops] update ci env
* [booster] optimize lora
* fix test
* fix test
2025-02-14 14:48:54 +08:00
Wenxuan Tan
ec73f1b5e2
[CI] Cleanup Dist Optim tests with shared helper funcs ( #6125 )
...
* Refractor and cleanup using common helper funcs. Tests passed
* Update comments
* Fix relative import
* Fix param fetching bug
2025-02-12 13:42:34 +08:00
flybird11111
5c09d726a6
[checkpointio] fix checkpoint for 3d ( #6187 )
...
* fix checkpoint io for 3d
* fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update hybrid_parallel_checkpoint_io.py
* fix
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-02-12 11:54:55 +08:00
Hongxin Liu
2b415e5999
[shardformer] support ep for deepseek v3 ( #6185 )
...
* [feature] support ep for deepseek v3
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix test
* [shardformer] fix deepseek v3 init
* [lazy] fit lora for lazy init
* [example] support npu for deepseek v3
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-02-11 16:10:25 +08:00
flybird11111
17062c83b9
[hotfix] fix hybrid checkpointio for sp+dp ( #6184 )
...
* Update hybrid_parallel_plugin.py
* Update hybrid_parallel_plugin.py
* Update hybrid_parallel_plugin.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update build_on_pr.yml
* Update test_zerobubble_pp.py
* fix
* fix
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-02-06 17:21:04 +08:00
Wenxuan Tan
ca0aa2365d
[Issue template] Add checkbox asking for details to reproduce error ( #6104 )
...
* Add checkbox asking about reproducing error
* update
* Update
* Update checkbox
2025-01-24 14:36:25 +08:00
Lemon Qin
97e60cbbcb
[checkpointio] gather tensor before unpad it if the tensor is both padded and distributed ( #6168 )
2025-01-21 10:23:15 +08:00
Guangyao Zhang
5b094a836b
[Inference]Fix example in readme ( #6178 )
2025-01-08 11:51:50 +08:00
Hongxin Liu
ee81366cac
[checkpointio] support load-pin overlap ( #6177 )
...
* [checkpointio] support load-pin overlap
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* [test] add conftest
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2025-01-07 16:16:04 +08:00
Hongxin Liu
479067e9bc
[release] update version ( #6174 )
...
* [release] update version
* [devops] fix test pypi ci
* [devops] fix test pypi ci
2025-01-03 11:52:23 +08:00
