d4fb7bfda7
fix typo applications/Chat/coati/ ( #3947 )
2023-06-15 10:43:11 +08:00
2925f47399
[evaluate] support gpt evaluation with reference ( #3972 )
...
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-06-13 15:12:29 +08:00
9d02590c9a
[chat] refactor actor class ( #3968 )
...
* refactor: separate log_probs fn from Actor forward fn
* refactor: separate generate fn from Actor class
* feat: update unwrap_model and get_base_model
* unwrap_model returns model not wrapped by Strategy
* get_base_model returns HF model for Actor, Critic and RewardModel
* feat: simplify Strategy.prepare
* style: remove get_base_model method of Actor
* perf: tokenize text in batches
* refactor: move calc_action_log_probs to utils of model
* test: update test with new forward fn
* style: rename forward fn args
* fix: do not unwrap model in save_model fn of naive strategy
* test: add gemini test for train_prompts
* fix: fix _set_default_generate_kwargs
2023-06-13 13:31:56 +08:00
21c4c0b1a0
support UniEval and add CHRF metric ( #3924 )
...
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-06-08 17:38:47 +08:00
b5f0566363
[chat] add distributed PPO trainer ( #3740 )
...
* Detached ppo (#9 )
* run the base
* working on dist ppo
* sync
* detached trainer
* update detached trainer. no maker update function
* facing init problem
* 1 maker 1 trainer detached run. but no model update
* facing cuda problem
* fix save functions
* verified maker update
* nothing
* add ignore
* analyize loss issue
* remove some debug codes
* facing 2m1t stuck issue
* 2m1t verified
* do not use torchrun
* working on 2m2t
* working on 2m2t
* initialize strategy in ray actor env
* facing actor's init order issue
* facing ddp model update issue (need unwarp ddp)
* unwrap ddp actor
* checking 1m2t stuck problem
* nothing
* set timeout for trainer choosing. It solves the stuck problem!
* delete some debug output
* rename to sync with upstream
* rename to sync with upstream
* coati rename
* nothing
* I am going to detach the replaybuffer from trainer and make it a Ray Actor. Two benefits: 1. support TP trainer. 2. asynchronized buffer operations
* experience_maker_holder performs target-revolving _send_experience() instead of length comparison.
* move code to ray subfolder
* working on pipeline inference
* apply comments
* working on pipeline strategy. in progress.
* remove pipeline code. clean this branch
* update remote parameters by state_dict. no test
* nothing
* state_dict sharding transfer
* merge debug branch
* gemini _unwrap_model fix
* simplify code
* simplify code & fix LoRALinear AttributeError
* critic unwrapped state_dict
---------
Co-authored-by: csric <richcsr256@gmail.com>
* [chat] add perfomance evaluator and fix bugs (#10 )
* [chat] add performance evaluator for ray
* [chat] refactor debug arg
* [chat] support hf config
* [chat] fix generation
* [chat] add 1mmt dummy example
* [chat] fix gemini ckpt
* split experience to send (#11 )
Co-authored-by: csric <richcsr256@gmail.com>
* [chat] refactor trainer and maker (#12 )
* [chat] refactor experience maker holder
* [chat] refactor model init
* [chat] refactor trainer args
* [chat] refactor model init
* [chat] refactor trainer
* [chat] refactor experience sending logic and training loop args (#13 )
* [chat] refactor experience send logic
* [chat] refactor trainer
* [chat] refactor trainer
* [chat] refactor experience maker
* [chat] refactor pbar
* [chat] refactor example folder (#14 )
* [chat] support quant (#15 )
* [chat] add quant
* [chat] add quant example
* prompt example (#16 )
* prompt example
* prompt load csv data
* remove legacy try
---------
Co-authored-by: csric <richcsr256@gmail.com>
* [chat] add mmmt dummy example and refactor experience sending (#17 )
* [chat] add mmmt dummy example
* [chat] refactor naive strategy
* [chat] fix struck problem
* [chat] fix naive strategy
* [chat] optimize experience maker sending logic
* [chat] refactor sending assignment
* [chat] refactor performance evaluator (#18 )
* Prompt Example & requires_grad state_dict & sharding state_dict (#19 )
* prompt example
* prompt load csv data
* remove legacy try
* maker models require_grad set to False
* working on zero redundancy update
* mmmt_prompt example; naive strategy requires_grad state_dict & sharding; maker model requires_no_grad.
* remove legacy examples
* remove legacy examples
* remove replay buffer tp state. bad design
---------
Co-authored-by: csric <richcsr256@gmail.com>
* state_dict sending adapts to new unwrap function (#20 )
* prompt example
* prompt load csv data
* remove legacy try
* maker models require_grad set to False
* working on zero redundancy update
* mmmt_prompt example; naive strategy requires_grad state_dict & sharding; maker model requires_no_grad.
* remove legacy examples
* remove legacy examples
* remove replay buffer tp state. bad design
* opt benchmark
* better script
* nothing
* [chat] strategy refactor unwrap model
* [chat] strategy refactor save model
* [chat] add docstr
* [chat] refactor trainer save model
* [chat] fix strategy typing
* [chat] refactor trainer save model
* [chat] update readme
* [chat] fix unit test
* working on lora reconstruction
* state_dict sending adapts to new unwrap function
* remove comments
---------
Co-authored-by: csric <richcsr256@gmail.com>
Co-authored-by: ver217 <lhx0217@gmail.com>
* [chat-ray] add readme (#21 )
* add readme
* transparent graph
* add note background
---------
Co-authored-by: csric <richcsr256@gmail.com>
* [chat] get images from url (#22 )
* Refactor/chat ray (#23 )
* [chat] lora add todo
* [chat] remove unused pipeline strategy
* [chat] refactor example structure
* [chat] setup ci for ray
* [chat-ray] Support LoRA trainer. LoRA weights reconstruction. (#24 )
* lora support prototype
* lora support
* 1mmt lora & remove useless code
---------
Co-authored-by: csric <richcsr256@gmail.com>
* [chat] fix test ci for ray
* [chat] fix test ci requirements for ray
* [chat] fix ray runtime env
* [chat] fix ray runtime env
* [chat] fix example ci docker args
* [chat] add debug info in trainer
* [chat] add nccl debug info
* [chat] skip ray test
* [doc] fix typo
---------
Co-authored-by: csric <59389055+CsRic@users.noreply.github.com>
Co-authored-by: csric <richcsr256@gmail.com>
2023-06-07 10:41:16 +08:00
57a6d7685c
support evaluation for english ( #3880 )
...
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-06-05 21:24:21 +08:00
2506e275b8
[evaluation] improvement on evaluation ( #3862 )
...
* fix a bug when the config file contains one category but the answer file doesn't contains that category
* fix Chinese prompt file
* support gpt-3.5-turbo and gpt-4 evaluation
* polish and update README
* resolve pr comments
---------
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-05-30 11:48:41 +08:00
e2d81eba0d
[nfc] fix typo colossalai/ applications/ ( #3831 )
...
* fix typo colossalai/autochunk auto_parallel amp
* fix typo colossalai/auto_parallel nn utils etc.
* fix typo colossalai/auto_parallel autochunk fx/passes etc.
* fix typo docs/
* change placememt_policy to placement_policy in docs/ and examples/
* fix typo colossalai/ applications/
2023-05-25 16:19:41 +08:00
34966378e8
[evaluation] add automatic evaluation pipeline ( #3821 )
...
* add functions for gpt evaluation
* add automatic eval
Update eval.py
* using jload and modify the type of answers1 and answers2
* Update eval.py
Update eval.py
* Update evaluator.py
* support gpt evaluation
* update readme.md
update README.md
update READNE.md
modify readme.md
* add Chinese example for config, battle prompt and evaluation prompt file
* remove GPT-4 config
* remove sample folder
---------
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
Co-authored-by: Camille Zhong <44392324+Camille7777@users.noreply.github.com>
2023-05-24 11:18:23 +08:00
9265f2d4d7
[NFC]fix typo colossalai/auto_parallel nn utils etc. ( #3779 )
...
* fix typo colossalai/autochunk auto_parallel amp
* fix typo colossalai/auto_parallel nn utils etc.
2023-05-23 15:28:20 +08:00
62c7e67f9f
[format] applied code formatting on changed files in pull request 3786 ( #3787 )
...
Co-authored-by: github-actions <github-actions@github.com>
2023-05-22 14:42:09 +08:00
ad2cf58f50
[chat] add performance and tutorial ( #3786 )
2023-05-19 18:03:56 +08:00
05759839bd
[chat] fix bugs in stage 3 training ( #3759 )
...
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-05-17 17:44:05 +08:00
ad6460cf2c
[NFC] fix typo applications/ and colossalai/ ( #3735 )
2023-05-15 11:46:25 +08:00
b7141c36dd
[CI] fix some spelling errors ( #3707 )
...
* fix spelling error with examples/comminity/
* fix spelling error with tests/
* fix some spelling error with tests/ colossalai/ etc.
2023-05-10 17:12:03 +08:00
f7361ee1bd
[chat] fix community example ray ( #3719 )
...
Co-authored-by: jiangwen <zxl265370@antgroup.com>
2023-05-10 13:36:09 +08:00
2da5d81dec
[chat] fix train_prompts.py gemini strategy bug ( #3666 )
...
* fix gemini strategy bug
* add comment
* add comment
* better solution
2023-05-06 16:46:38 +08:00
65bdc3159f
fix some spelling error with applications/Chat/examples/ ( #3692 )
...
* fix spelling error with examples/comminity/
* fix spelling error with example/
2023-05-06 11:27:23 +08:00
b36e67cb2b
Merge pull request #3680 from digger-yu/digger-yu-patch-2
...
fix spelling error with applications/Chat/evaluate/
2023-05-05 16:26:04 +08:00
0f785cb1f3
[chat] PPO stage3 doc enhancement ( #3679 )
...
* Add RoBERTa for RLHF Stage 2 & 3 (test)
RoBERTa for RLHF Stage 2 & 3 (still in testing)
Revert "Add RoBERTa for RLHF Stage 2 & 3 (test)"
This reverts commit 06741d894d06741d894d
2023-05-05 13:36:56 +08:00
6650daeb0a
[doc] fix chat spelling error ( #3671 )
...
* Update README.md
change "huggingaface" to "huggingface"
* Update README.md
change "Colossa-AI" to "Colossal-AI"
2023-05-05 11:37:35 +08:00
7bd0bee8ea
[chat] add opt attn kernel ( #3655 )
...
* [chat] add opt attn kernel
* [chat] disable xformer during fwd
2023-05-04 16:03:33 +08:00
8ba7858753
Update generate_gpt35_answers.py
...
fix spelling error with generate_gpt35_answers.py
2023-05-04 15:34:16 +08:00
bfbf650588
fix spelling error
...
fix spelling error with evaluate.py
2023-05-04 15:31:09 +08:00
1a60dc07a8
[chat] typo accimulation_steps -> accumulation_steps ( #3662 )
2023-04-28 15:42:57 +08:00
816add7e7f
Merge pull request #3656 from TongLi3701/chat/update_eval
...
[Chat]: Remove unnecessary step and update documentation
2023-04-28 14:07:44 +08:00
268b3cd80d
[chat] set default zero2 strategy ( #3667 )
...
* [chat] set default gemini strategy
* [chat] set default zero2 strategy
* [chat] set default zero2 strategy
2023-04-28 13:56:50 +08:00
c1a355940e
update readme
2023-04-28 11:56:35 +08:00
ed3eaa6922
update documentation
2023-04-28 11:49:21 +08:00
c419117329
update questions and readme
2023-04-27 19:04:26 +08:00
aa77ddae33
remove unnecessary step and update readme
2023-04-27 18:51:58 +08:00
842768a174
[chat] refactor model save/load logic ( #3654 )
...
* [chat] strategy refactor unwrap model
* [chat] strategy refactor save model
* [chat] add docstr
* [chat] refactor trainer save model
* [chat] fix strategy typing
* [chat] refactor trainer save model
* [chat] update readme
* [chat] fix unit test
2023-04-27 18:41:49 +08:00
6ef7011462
[chat] remove lm model class ( #3653 )
...
* [chat] refactor lora
* [chat] remove lm class
* [chat] refactor save model
* [chat] refactor train sft
* [chat] fix ci
* [chat] fix ci
2023-04-27 15:37:38 +08:00
8bccb72c8d
[Doc] enhancement on README.md for chat examples ( #3646 )
...
* Add RoBERTa for RLHF Stage 2 & 3 (test)
RoBERTa for RLHF Stage 2 & 3 (still in testing)
Revert "Add RoBERTa for RLHF Stage 2 & 3 (test)"
This reverts commit 06741d894d06741d894d
2023-04-27 14:26:19 +08:00
2a951955ad
[chat] refactor trainer ( #3648 )
...
* [chat] ppo trainer remove useless args
* [chat] update examples
* [chat] update benchmark
* [chat] update examples
* [chat] fix sft training with wandb
* [chat] polish docstr
2023-04-26 18:11:49 +08:00
f8288315d9
[chat] polish performance evaluator ( #3647 )
2023-04-26 17:34:59 +08:00
50793b35f4
[gemini] accelerate inference ( #3641 )
...
* [gemini] support don't scatter after inference
* [chat] update colossalai strategy
* [chat] fix opt benchmark
* [chat] update opt benchmark
* [gemini] optimize inference
* [test] add gemini inference test
* [chat] fix unit test ci
* [chat] fix ci
* [chat] fix ci
* [chat] skip checkpoint test
2023-04-26 16:32:40 +08:00
e1b0a78afa
Merge pull request #3621 from zhang-yi-chi/fix/chat-train-prompts-single-gpu
...
[chat] fix single gpu training bug in examples/train_prompts.py
2023-04-24 22:13:54 +08:00
df309fc6ab
[Chat] Remove duplicate functions ( #3625 )
2023-04-24 12:23:15 +08:00
739cfe3360
[chat] fix enable single gpu training bug
2023-04-22 14:16:08 +08:00
d7bf284706
[chat] polish code note typo ( #3612 )
2023-04-20 17:22:15 +08:00
c4709d34cf
Chat evaluate ( #3608 )
...
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-04-20 11:12:24 +08:00
5a79cffdfd
[coati] fix install cmd ( #3592 )
2023-04-18 18:19:48 +08:00
1ec0d386a9
reconstruct chat trainer and fix training script ( #3588 )
...
Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>
2023-04-18 16:44:03 +08:00
36a519b49f
Update test_ci.sh
...
update
Update test_ci.sh
Update test_ci.sh
Update test_ci.sh
Update test_ci.sh
Update test_ci.sh
Update test_ci.sh
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update test_ci.sh
Update test_ci.sh
update
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
update ci
Update test_ci.sh
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update run_chatgpt_examples.yml
Update test_ci.sh
Update test_ci.sh
Update run_chatgpt_examples.yml
Update test_ci.sh
Update test_ci.sh
Update test_ci.sh
update test ci
RoBERTa for RLHF Stage 2 & 3 (still in testing)
Revert "Add RoBERTa for RLHF Stage 2 & 3 (test)"
This reverts commit 06741d894d06741d894d
2023-04-18 14:33:12 +08:00
7788e0b0a5
fix: fix sft ( #3568 )
2023-04-17 16:47:44 +08:00
6b1a39b17b
[coati] add costom model suppor tguide ( #3579 )
2023-04-17 15:40:41 +08:00
cc1eec2f53
[chat] update reward model sh ( #3578 )
2023-04-17 15:02:55 +08:00
e355144375
[chatgpt] Detached PPO Training ( #3195 )
...
* run the base
* working on dist ppo
* sync
* detached trainer
* update detached trainer. no maker update function
* facing init problem
* 1 maker 1 trainer detached run. but no model update
* facing cuda problem
* fix save functions
* verified maker update
* nothing
* add ignore
* analyize loss issue
* remove some debug codes
* facing 2m1t stuck issue
* 2m1t verified
* do not use torchrun
* working on 2m2t
* working on 2m2t
* initialize strategy in ray actor env
* facing actor's init order issue
* facing ddp model update issue (need unwarp ddp)
* unwrap ddp actor
* checking 1m2t stuck problem
* nothing
* set timeout for trainer choosing. It solves the stuck problem!
* delete some debug output
* rename to sync with upstream
* rename to sync with upstream
* coati rename
* nothing
* I am going to detach the replaybuffer from trainer and make it a Ray Actor. Two benefits: 1. support TP trainer. 2. asynchronized buffer operations
* experience_maker_holder performs target-revolving _send_experience() instead of length comparison.
* move code to ray subfolder
* working on pipeline inference
* apply comments
---------
Co-authored-by: csric <richcsr256@gmail.com>
2023-04-17 14:46:50 +08:00
1a809eddaa
[chat] ChatGPT train prompts on ray example ( #3309 )
...
* [feat][chatgpt]train prompts on ray example
* [fix]simplify code
* [fix]remove depreciated parameter
* [fix]add dependencies
* [fix]method calling
* [fix]experience maker
* [fix]missing loss function
* [fix]init optimizer
* [feat]add usage comment
* [fix]rename files
* [fix]add readme
* [fix]file path
* [fix]move directory
---------
Co-authored-by: jiangwen <zxl265370@antgroup.com>
2023-04-13 18:18:36 +08:00
digger yu