[shardformer]enable padding vocabulary size. #5489

flybird11111 · 2024-03-22T03:37:30Z

📌 Checklist before creating the PR

I have created an issue for this PR for traceability
The title follows the standard format: [doc/gemini/tensor/...]: A concise description
I have added relevant tags if possible for us to better distinguish different PRs

🚨 Issue number

Link this PR to your issue with words like fixed to automatically close the linked issue upon merge

e.g. fixed #1234, closed #1234, resolved #1234

#5488

📝 What does this PR do?

Summarize your work here.
if you have any plots/diagrams/screenshots/tables, please attach them here.

enable padding vocabulary size to make it choose a faster kernel.

💥 Checklist before requesting a review

I have linked my PR to an issue (instruction)
My issue clearly describes the problem/feature/proposal, with diagrams/charts/table/code if possible
I have performed a self-review of my code
I have added thorough tests.
I have added docstrings for all the functions/methods I implemented

⭐️ Do you enjoy contributing to Colossal-AI?

🌝 Yes, I do.
🌚 No, I don't.

Tell us more if you don't enjoy contributing to Colossal-AI.

flybird11111 · 2024-03-29T09:14:19Z

All tests have passed.

colossalai/shardformer/layer/parallel_module.py

colossalai/shardformer/layer/linear.py

colossalai/shardformer/layer/parallel_module.py

colossalai/shardformer/policies/bert.py

tests/test_shardformer/test_model/_utils.py

tests/test_optimizer/test_nvme.py

tests/test_checkpoint_io/test_gemini_checkpoint_io.py

flybird11111 · 2024-04-02T11:17:22Z

All tests have passed.

colossalai/shardformer/layer/embedding.py

colossalai/shardformer/layer/linear.py

flybird11111 · 2024-04-03T10:57:20Z

All tests have passed.

colossalai/shardformer/layer/linear.py

insujang · 2024-04-08T13:17:30Z

Hi @ver217 @flybird11111 , thank you for the great work. I recent found a bug in hybrid parallel execution (#5539) that HybridParallelOptimizer holds unsharded parameters, and it was because of vocab embedidng padding (in each model policy's preprocess, calling self.model.resize_token_embeddings() creates a new nn.Embedding leaving Optimizer to look the old embedding parameter.
Since this function call has been removed in this PR, I wonder if it solves my problem.

colossalai/shardformer/layer/parallel_module.py

flybird11111 · 2024-04-09T05:49:48Z

Hi @ver217 @flybird11111 , thank you for the great work. I recent found a bug in hybrid parallel execution (#5539) that HybridParallelOptimizer holds unsharded parameters, and it was because of vocab embedidng padding (in each model policy's preprocess, calling self.model.resize_token_embeddings() creates a new nn.Embedding leaving Optimizer to look the old embedding parameter. Since this function call has been removed in this PR, I wonder if it solves my problem.

Yes, this pull request can resolve your problem.

for more information, see https://pre-commit.ci

fix fix fix fix fix fix fix

colossalai/booster/plugin/gemini_plugin.py

colossalai/checkpoint_io/hybrid_parallel_checkpoint_io.py

colossalai/booster/plugin/gemini_plugin.py

colossalai/zero/gemini/gemini_ddp.py

tests/test_checkpoint_io/test_gemini_checkpoint_io.py

resolve comments resolve comments resolve comments resolve comments

colossalai/checkpoint_io/hybrid_parallel_checkpoint_io.py

colossalai/tensor/p_tensor/__init__.py

colossalai/tensor/p_tensor/api.py

tests/test_checkpoint_io/test_hybrid_parallel_plugin_checkpoint_io.py

tests/test_shardformer/test_model/_utils.py

tests/test_tensor/test_padded_tensor/test_padded_tensor.py

ptensor resolve comments fix fix fix fix fix resolve comments resolve comments resolve comments resolve comments resolve comments

* [branch rebase] rebase main to Feature/resize_embedding (#5554) * fix * [release] update version (#5411) * [hotfix] fix typo s/keywrods/keywords etc. (#5429) * [devops] fix compatibility (#5444) * [devops] fix compatibility * [hotfix] update compatibility test on pr * [devops] fix compatibility * [devops] record duration during comp test * [test] decrease test duration * fix falcon * [shardformer] fix gathering output when using tensor parallelism (#5431) * fix * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * [doc] release Open-Sora 1.0 with model weights (#5468) * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] update open-sora demo (#5479) * [doc] update open-sora demo * [doc] update open-sora demo * [doc] update open-sora demo * [example] add grok-1 inference (#5485) * [misc] add submodule * remove submodule * [example] support grok-1 tp inference * [example] add grok-1 inference script * [example] refactor code * [example] add grok-1 readme * [exmaple] add test ci * [exmaple] update readme --------- Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: binmakeswell <[email protected]> * [CI] run pre-commit (#5577) * fix * [release] update version (#5411) * [hotfix] fix typo s/keywrods/keywords etc. (#5429) * [devops] fix compatibility (#5444) * [devops] fix compatibility * [hotfix] update compatibility test on pr * [devops] fix compatibility * [devops] record duration during comp test * [test] decrease test duration * fix falcon * [shardformer] fix gathering output when using tensor parallelism (#5431) * fix * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * [doc] release Open-Sora 1.0 with model weights (#5468) * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] update open-sora demo (#5479) * [doc] update open-sora demo * [doc] update open-sora demo * [doc] update open-sora demo * [example] add grok-1 inference (#5485) * [misc] add submodule * remove submodule * [example] support grok-1 tp inference * [example] add grok-1 inference script * [example] refactor code * [example] add grok-1 readme * [exmaple] add test ci * [exmaple] update readme * run pre-commit --------- Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: binmakeswell <[email protected]> * [rebase] rebase main to resize-embedding (#5581) * [release] grok-1 314b inference (#5490) * [release] grok-1 inference * [release] grok-1 inference * [release] grok-1 inference * [example] update Grok-1 inference (#5495) * revise grok-1 example * remove unused arg in scripts * prevent re-installing torch * update readme * revert modifying colossalai requirements * add perf * trivial * add tokenizer url * [hotfix] set return_outputs=False in examples and polish code (#5404) * fix: simplify merge_batch * fix: use return_outputs=False to eliminate extra memory consumption * feat: add return_outputs warning * style: remove `return_outputs=False` as it is the default value * [release] grok-1 inference benchmark (#5500) * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [shardformer]Fix lm parallel. (#5480) * fix * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * fix lm forward distribution * fix * test ci * fix * [fix] fix grok-1 example typo (#5506) * [devops] fix example test ci (#5504) * Fix ColoTensorSpec for py11 (#5440) * fixed layout converter caching and updated tester * Empty-Commit * [shardformer] update colo attention to support custom mask (#5510) * [feature] refactor colo attention (#5462) * [extension] update api * [feature] add colo attention * [feature] update sdpa * [feature] update npu attention * [feature] update flash-attn * [test] add flash attn test * [test] update flash attn test * [shardformer] update modeling to fit colo attention (#5465) * [misc] refactor folder structure * [shardformer] update llama flash-attn * [shardformer] fix llama policy * [devops] update tensornvme install * [test] update llama test * [shardformer] update colo attn kernel dispatch * [shardformer] update blip2 * [shardformer] update chatglm * [shardformer] update gpt2 * [shardformer] update gptj * [shardformer] update opt * [shardformer] update vit * [shardformer] update colo attention mask prep * [shardformer] update whisper * [test] fix shardformer tests (#5514) * [test] fix shardformer tests * [test] fix shardformer tests * [format] applied code formatting on changed files in pull request 5510 (#5517) Co-authored-by: github-actions <[email protected]> * [shardformer] fix pipeline forward error if custom layer distribution is used (#5189) * Use self.[distribute_layers|get_stage_index] to exploit custom layer distribution * Change static methods for t5 layer distribution to member functions * Change static methods for whisper layer distribution to member functions * Replace whisper policy usage with self one * Fix test case to use non-static layer distribution methods * fix: fix typo --------- Co-authored-by: Wenhao Chen <[email protected]> * [Fix] Grok-1 use tokenizer from the same pretrained path (#5532) * [fix] use tokenizer from the same pretrained path * trust remote code * [ColossalChat] Update RLHF V2 (#5286) * Add dpo. Fix sft, ppo, lora. Refactor all * fix and tested ppo * 2 nd round refactor * add ci tests * fix ci * fix ci * fix readme, style * fix readme style * fix style, fix benchmark * reproduce benchmark result, remove useless files * rename to ColossalChat * use new image * fix ci workflow * fix ci * use local model/tokenizer for ci tests * fix ci * fix ci * fix ci * fix ci timeout * fix rm progress bar. fix ci timeout * fix ci * fix ci typo * remove 3d plugin from ci temporary * test environment * cannot save optimizer * support chat template * fix readme * fix path * test ci locally * restore build_or_pr * fix ci data path * fix benchmark * fix ci, move ci tests to 3080, disable fast tokenizer * move ci to 85 * support flash attention 2 * add all-in-one data preparation script. Fix colossal-llama2-chat chat template * add hardware requirements * move ci test data * fix save_model, add unwrap * fix missing bos * fix missing bos; support grad accumulation with gemini * fix ci * fix ci * fix ci * fix llama2 chat template config * debug sft * debug sft * fix colossalai version requirement * fix ci * add sanity check to prevent NaN loss * fix requirements * add dummy data generation script * add dummy data generation script * add dummy data generation script * add dummy data generation script * update readme * update readme * update readme and ignore * fix logger bug * support parallel_output * modify data preparation logic * fix tokenization * update lr * fix inference * run pre-commit --------- Co-authored-by: Tong Li <[email protected]> * [shardformer, pipeline] add `gradient_checkpointing_ratio` and heterogenous shard policy for llama (#5508) * feat: add `GradientCheckpointConfig` and `PipelineGradientCheckpointConfig` * feat: apply `GradientCheckpointConfig` to policy and llama_forward * feat: move `distribute_layer` and `get_stage_index` to PipelineStageManager * fix: add optional args for `distribute_layer` and `get_stage_index` * fix: fix changed API calls * test: update llama tests * style: polish `GradientCheckpointConfig` * fix: fix pipeline utils tests * fix incorrect sharding without zero (#5545) Co-authored-by: Edenzzzz <[email protected]> * [shardformer] Sequence Parallelism Optimization (#5533) * sequence parallel optimization * validate sequence parallel in llama (code to be polished) * shardformer api writing * integrate sequence parallel in ShardFormer * fix pp bugs and sp bugs for LlaMa model * integrating ring-based sequence parallelism into ShardFormer * [sequence parallelism]: Add fused megatron function * integrating ring-based sequence parallelism into ShardFormer --------- Co-authored-by: linsj20 <[email protected]> * fix bugs when useing sp and flashattention together * fix operation function name * support flash attention for ulysses-style sp * clarify sp process group * fix compatibility bugs in moe plugin * fix fused linear bugs * fix linear layer test * support gpt model all-to-all sp * modify shard data dimension (meant to be dim=-1) * support megtron-style sp and distributed attn for llama model * [shardformer] add megatron sp to llama * support llama7B 128k with distributed attention * [shardformer] robustness enhancement * add block attn * sp mode 1: keep input as a complete sequence * fix sp compatability * finish sp mode 3 support for gpt * using all_to_all_single when batch size is 1 * support mode 2 sp in gpt2 (#5) * [shardformer] add megatron sp to llama * support llama7B 128k with distributed attention * [shardformer] robustness enhancement * add block attn * sp mode 1: keep input as a complete sequence * fix sp compatability * refactor ring implementation * support mode 2 sp in gpt2 * polish code * enable distributed attn mask when using sp mode 2 and 3 in llama * automatically enable flash attn when using sp mode 2 and 3 in llama * inplace attn mask * add zero2 support for sequence parallel * polish code * fix bugs * fix gemini checkpoint io * loose tensor checking atol and rtol * add comment * fix llama layernorm grad * fix zero grad * fix zero grad * fix conflict * update split and gather auto grad func * sequence parallel: inside text split (#6) * polish code (part 1) * polish code (part 2) * polish code (part 2.5) * polish code (part 3) * sequence parallel: inside text split * miscellaneous minor fixes * polish code * fix ulysses style ZeRO * sequence parallel: inside text split * miscellaneous minor fixes * disaggregate sp group and dp group for sp * fix llama and gpt sp * polish code * move ulysses grad sync to ddp (#9) * remove zero_stage and unbind the grad sync for alltoall sp * add 2d group creation test * move ulysses grad sync to ddp * add 2d group creation test * remove useless code * change shard config not to enable sp when enable_all_optimizations * add sp warnings for several model * remove useless code --------- Co-authored-by: linsj20 <[email protected]> * [hotfix] quick fixes to make legacy tutorials runnable (#5559) Co-authored-by: Edenzzzz <[email protected]> * [fix] fix typo s/muiti-node /multi-node etc. (#5448) * [hotfix] fix typo s/get_defualt_parser /get_default_parser (#5548) * [devops] remove post commit ci (#5566) * [devops] remove post commit ci * [misc] run pre-commit on all files * [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> --------- Co-authored-by: binmakeswell <[email protected]> Co-authored-by: Yuanheng Zhao <[email protected]> Co-authored-by: Wenhao Chen <[email protected]> Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: Rocky Duan <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions <[email protected]> Co-authored-by: Insu Jang <[email protected]> Co-authored-by: YeAnbang <[email protected]> Co-authored-by: Tong Li <[email protected]> Co-authored-by: Zhongkai Zhao <[email protected]> Co-authored-by: linsj20 <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * [shardformer]enable padding vocabulary size. (#5489) * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * padding vocab * padding vocabe * fix * fix * fxi * test ci * fix fix fix fix * fix fix * fix * fix * Update hybrid_parallel_plugin.py fix fix fix * fix fix * fix fix * fix * resolve super init resolve super init resolve super init resolve super init * resolve comments * fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * vocab checkpointio * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix fix fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * padding vocab * fix * fix fix * fix fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix ci * fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix * cherry-pick * revert moe modify * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix fix fix fix fix fix fix fix * resolve comments resolve comments resolve comments resolve comments resolve comments * ptensor ptensor resolve comments fix fix fix fix fix resolve comments resolve comments resolve comments resolve comments resolve comments --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Hongxin Liu <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix rebase * fix rebase --------- Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: binmakeswell <[email protected]> Co-authored-by: Yuanheng Zhao <[email protected]> Co-authored-by: Wenhao Chen <[email protected]> Co-authored-by: Rocky Duan <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions <[email protected]> Co-authored-by: Insu Jang <[email protected]> Co-authored-by: YeAnbang <[email protected]> Co-authored-by: Tong Li <[email protected]> Co-authored-by: Zhongkai Zhao <[email protected]> Co-authored-by: linsj20 <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

* [branch rebase] rebase main to Feature/resize_embedding (hpcaitech#5554) * fix * [release] update version (hpcaitech#5411) * [hotfix] fix typo s/keywrods/keywords etc. (hpcaitech#5429) * [devops] fix compatibility (hpcaitech#5444) * [devops] fix compatibility * [hotfix] update compatibility test on pr * [devops] fix compatibility * [devops] record duration during comp test * [test] decrease test duration * fix falcon * [shardformer] fix gathering output when using tensor parallelism (hpcaitech#5431) * fix * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * [doc] release Open-Sora 1.0 with model weights (hpcaitech#5468) * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] update open-sora demo (hpcaitech#5479) * [doc] update open-sora demo * [doc] update open-sora demo * [doc] update open-sora demo * [example] add grok-1 inference (hpcaitech#5485) * [misc] add submodule * remove submodule * [example] support grok-1 tp inference * [example] add grok-1 inference script * [example] refactor code * [example] add grok-1 readme * [exmaple] add test ci * [exmaple] update readme --------- Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: binmakeswell <[email protected]> * [CI] run pre-commit (hpcaitech#5577) * fix * [release] update version (hpcaitech#5411) * [hotfix] fix typo s/keywrods/keywords etc. (hpcaitech#5429) * [devops] fix compatibility (hpcaitech#5444) * [devops] fix compatibility * [hotfix] update compatibility test on pr * [devops] fix compatibility * [devops] record duration during comp test * [test] decrease test duration * fix falcon * [shardformer] fix gathering output when using tensor parallelism (hpcaitech#5431) * fix * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * [doc] release Open-Sora 1.0 with model weights (hpcaitech#5468) * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] release Open-Sora 1.0 with model weights * [doc] update open-sora demo (hpcaitech#5479) * [doc] update open-sora demo * [doc] update open-sora demo * [doc] update open-sora demo * [example] add grok-1 inference (hpcaitech#5485) * [misc] add submodule * remove submodule * [example] support grok-1 tp inference * [example] add grok-1 inference script * [example] refactor code * [example] add grok-1 readme * [exmaple] add test ci * [exmaple] update readme * run pre-commit --------- Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: binmakeswell <[email protected]> * [rebase] rebase main to resize-embedding (hpcaitech#5581) * [release] grok-1 314b inference (hpcaitech#5490) * [release] grok-1 inference * [release] grok-1 inference * [release] grok-1 inference * [example] update Grok-1 inference (hpcaitech#5495) * revise grok-1 example * remove unused arg in scripts * prevent re-installing torch * update readme * revert modifying colossalai requirements * add perf * trivial * add tokenizer url * [hotfix] set return_outputs=False in examples and polish code (hpcaitech#5404) * fix: simplify merge_batch * fix: use return_outputs=False to eliminate extra memory consumption * feat: add return_outputs warning * style: remove `return_outputs=False` as it is the default value * [release] grok-1 inference benchmark (hpcaitech#5500) * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [release] grok-1 inference benchmark * [shardformer]Fix lm parallel. (hpcaitech#5480) * fix * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * fix lm forward distribution * fix * test ci * fix * [fix] fix grok-1 example typo (hpcaitech#5506) * [devops] fix example test ci (hpcaitech#5504) * Fix ColoTensorSpec for py11 (hpcaitech#5440) * fixed layout converter caching and updated tester * Empty-Commit * [shardformer] update colo attention to support custom mask (hpcaitech#5510) * [feature] refactor colo attention (hpcaitech#5462) * [extension] update api * [feature] add colo attention * [feature] update sdpa * [feature] update npu attention * [feature] update flash-attn * [test] add flash attn test * [test] update flash attn test * [shardformer] update modeling to fit colo attention (hpcaitech#5465) * [misc] refactor folder structure * [shardformer] update llama flash-attn * [shardformer] fix llama policy * [devops] update tensornvme install * [test] update llama test * [shardformer] update colo attn kernel dispatch * [shardformer] update blip2 * [shardformer] update chatglm * [shardformer] update gpt2 * [shardformer] update gptj * [shardformer] update opt * [shardformer] update vit * [shardformer] update colo attention mask prep * [shardformer] update whisper * [test] fix shardformer tests (hpcaitech#5514) * [test] fix shardformer tests * [test] fix shardformer tests * [format] applied code formatting on changed files in pull request 5510 (hpcaitech#5517) Co-authored-by: github-actions <[email protected]> * [shardformer] fix pipeline forward error if custom layer distribution is used (hpcaitech#5189) * Use self.[distribute_layers|get_stage_index] to exploit custom layer distribution * Change static methods for t5 layer distribution to member functions * Change static methods for whisper layer distribution to member functions * Replace whisper policy usage with self one * Fix test case to use non-static layer distribution methods * fix: fix typo --------- Co-authored-by: Wenhao Chen <[email protected]> * [Fix] Grok-1 use tokenizer from the same pretrained path (hpcaitech#5532) * [fix] use tokenizer from the same pretrained path * trust remote code * [ColossalChat] Update RLHF V2 (hpcaitech#5286) * Add dpo. Fix sft, ppo, lora. Refactor all * fix and tested ppo * 2 nd round refactor * add ci tests * fix ci * fix ci * fix readme, style * fix readme style * fix style, fix benchmark * reproduce benchmark result, remove useless files * rename to ColossalChat * use new image * fix ci workflow * fix ci * use local model/tokenizer for ci tests * fix ci * fix ci * fix ci * fix ci timeout * fix rm progress bar. fix ci timeout * fix ci * fix ci typo * remove 3d plugin from ci temporary * test environment * cannot save optimizer * support chat template * fix readme * fix path * test ci locally * restore build_or_pr * fix ci data path * fix benchmark * fix ci, move ci tests to 3080, disable fast tokenizer * move ci to 85 * support flash attention 2 * add all-in-one data preparation script. Fix colossal-llama2-chat chat template * add hardware requirements * move ci test data * fix save_model, add unwrap * fix missing bos * fix missing bos; support grad accumulation with gemini * fix ci * fix ci * fix ci * fix llama2 chat template config * debug sft * debug sft * fix colossalai version requirement * fix ci * add sanity check to prevent NaN loss * fix requirements * add dummy data generation script * add dummy data generation script * add dummy data generation script * add dummy data generation script * update readme * update readme * update readme and ignore * fix logger bug * support parallel_output * modify data preparation logic * fix tokenization * update lr * fix inference * run pre-commit --------- Co-authored-by: Tong Li <[email protected]> * [shardformer, pipeline] add `gradient_checkpointing_ratio` and heterogenous shard policy for llama (hpcaitech#5508) * feat: add `GradientCheckpointConfig` and `PipelineGradientCheckpointConfig` * feat: apply `GradientCheckpointConfig` to policy and llama_forward * feat: move `distribute_layer` and `get_stage_index` to PipelineStageManager * fix: add optional args for `distribute_layer` and `get_stage_index` * fix: fix changed API calls * test: update llama tests * style: polish `GradientCheckpointConfig` * fix: fix pipeline utils tests * fix incorrect sharding without zero (hpcaitech#5545) Co-authored-by: Edenzzzz <[email protected]> * [shardformer] Sequence Parallelism Optimization (hpcaitech#5533) * sequence parallel optimization * validate sequence parallel in llama (code to be polished) * shardformer api writing * integrate sequence parallel in ShardFormer * fix pp bugs and sp bugs for LlaMa model * integrating ring-based sequence parallelism into ShardFormer * [sequence parallelism]: Add fused megatron function * integrating ring-based sequence parallelism into ShardFormer --------- Co-authored-by: linsj20 <[email protected]> * fix bugs when useing sp and flashattention together * fix operation function name * support flash attention for ulysses-style sp * clarify sp process group * fix compatibility bugs in moe plugin * fix fused linear bugs * fix linear layer test * support gpt model all-to-all sp * modify shard data dimension (meant to be dim=-1) * support megtron-style sp and distributed attn for llama model * [shardformer] add megatron sp to llama * support llama7B 128k with distributed attention * [shardformer] robustness enhancement * add block attn * sp mode 1: keep input as a complete sequence * fix sp compatability * finish sp mode 3 support for gpt * using all_to_all_single when batch size is 1 * support mode 2 sp in gpt2 (hpcaitech#5) * [shardformer] add megatron sp to llama * support llama7B 128k with distributed attention * [shardformer] robustness enhancement * add block attn * sp mode 1: keep input as a complete sequence * fix sp compatability * refactor ring implementation * support mode 2 sp in gpt2 * polish code * enable distributed attn mask when using sp mode 2 and 3 in llama * automatically enable flash attn when using sp mode 2 and 3 in llama * inplace attn mask * add zero2 support for sequence parallel * polish code * fix bugs * fix gemini checkpoint io * loose tensor checking atol and rtol * add comment * fix llama layernorm grad * fix zero grad * fix zero grad * fix conflict * update split and gather auto grad func * sequence parallel: inside text split (hpcaitech#6) * polish code (part 1) * polish code (part 2) * polish code (part 2.5) * polish code (part 3) * sequence parallel: inside text split * miscellaneous minor fixes * polish code * fix ulysses style ZeRO * sequence parallel: inside text split * miscellaneous minor fixes * disaggregate sp group and dp group for sp * fix llama and gpt sp * polish code * move ulysses grad sync to ddp (hpcaitech#9) * remove zero_stage and unbind the grad sync for alltoall sp * add 2d group creation test * move ulysses grad sync to ddp * add 2d group creation test * remove useless code * change shard config not to enable sp when enable_all_optimizations * add sp warnings for several model * remove useless code --------- Co-authored-by: linsj20 <[email protected]> * [hotfix] quick fixes to make legacy tutorials runnable (hpcaitech#5559) Co-authored-by: Edenzzzz <[email protected]> * [fix] fix typo s/muiti-node /multi-node etc. (hpcaitech#5448) * [hotfix] fix typo s/get_defualt_parser /get_default_parser (hpcaitech#5548) * [devops] remove post commit ci (hpcaitech#5566) * [devops] remove post commit ci * [misc] run pre-commit on all files * [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> --------- Co-authored-by: binmakeswell <[email protected]> Co-authored-by: Yuanheng Zhao <[email protected]> Co-authored-by: Wenhao Chen <[email protected]> Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: Rocky Duan <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions <[email protected]> Co-authored-by: Insu Jang <[email protected]> Co-authored-by: YeAnbang <[email protected]> Co-authored-by: Tong Li <[email protected]> Co-authored-by: Zhongkai Zhao <[email protected]> Co-authored-by: linsj20 <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * [shardformer]enable padding vocabulary size. (hpcaitech#5489) * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix * fix fix fix * fix gather output * fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * revert * padding vocab * padding vocabe * fix * fix * fxi * test ci * fix fix fix fix * fix fix * fix * fix * Update hybrid_parallel_plugin.py fix fix fix * fix fix * fix fix * fix * resolve super init resolve super init resolve super init resolve super init * resolve comments * fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * vocab checkpointio * padding vocab_size when using pipeline parallellism padding vocab_size when using pipeline parallellism fix fix * fix fix fix * fix * fix fix resize embedding fix resize embedding * fix resize embedding fix * revert * revert * padding vocab * fix * fix fix * fix fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix ci * fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix * cherry-pick * revert moe modify * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix fix fix fix fix fix fix fix * resolve comments resolve comments resolve comments resolve comments resolve comments * ptensor ptensor resolve comments fix fix fix fix fix resolve comments resolve comments resolve comments resolve comments resolve comments --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Hongxin Liu <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix rebase * fix rebase --------- Co-authored-by: Hongxin Liu <[email protected]> Co-authored-by: digger yu <[email protected]> Co-authored-by: binmakeswell <[email protected]> Co-authored-by: Yuanheng Zhao <[email protected]> Co-authored-by: Wenhao Chen <[email protected]> Co-authored-by: Rocky Duan <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: Edenzzzz <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions <[email protected]> Co-authored-by: Insu Jang <[email protected]> Co-authored-by: YeAnbang <[email protected]> Co-authored-by: Tong Li <[email protected]> Co-authored-by: Zhongkai Zhao <[email protected]> Co-authored-by: linsj20 <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

flybird11111 requested a review from a team as a code owner March 22, 2024 03:37

flybird11111 marked this pull request as draft March 22, 2024 03:47

flybird11111 marked this pull request as ready for review March 22, 2024 03:47

flybird11111 force-pushed the padding-vocab branch 7 times, most recently from be6625c to 4782565 Compare March 29, 2024 08:12

ver217 reviewed Apr 1, 2024

View reviewed changes

colossalai/shardformer/layer/parallel_module.py Show resolved Hide resolved

colossalai/shardformer/layer/linear.py Outdated Show resolved Hide resolved

colossalai/shardformer/layer/linear.py Outdated Show resolved Hide resolved

ver217 reviewed Apr 1, 2024

View reviewed changes

colossalai/shardformer/layer/linear.py Outdated Show resolved Hide resolved

colossalai/shardformer/layer/parallel_module.py Outdated Show resolved Hide resolved

colossalai/shardformer/policies/bert.py Outdated Show resolved Hide resolved

ver217 reviewed Apr 1, 2024

View reviewed changes

tests/test_shardformer/test_model/_utils.py Outdated Show resolved Hide resolved

tests/test_optimizer/test_nvme.py Outdated Show resolved Hide resolved

tests/test_checkpoint_io/test_gemini_checkpoint_io.py Outdated Show resolved Hide resolved

flybird11111 force-pushed the padding-vocab branch 2 times, most recently from 51665a2 to 8f2db47 Compare April 2, 2024 10:04

ver217 reviewed Apr 3, 2024

View reviewed changes

colossalai/shardformer/layer/embedding.py Outdated Show resolved Hide resolved

colossalai/shardformer/layer/embedding.py Outdated Show resolved Hide resolved

colossalai/shardformer/layer/linear.py Outdated Show resolved Hide resolved

flybird11111 force-pushed the padding-vocab branch 4 times, most recently from 84cbb6a to ed4a808 Compare April 3, 2024 10:56

ver217 reviewed Apr 8, 2024

View reviewed changes

colossalai/shardformer/layer/linear.py Outdated Show resolved Hide resolved

insujang mentioned this pull request Apr 8, 2024

[BUG]: HybridParallelOptimizer holds unsharded model parameters after sharding #5539

Closed

insujang reviewed Apr 8, 2024

View reviewed changes

colossalai/shardformer/layer/parallel_module.py Outdated Show resolved Hide resolved

flybird11111 force-pushed the padding-vocab branch from 86a115d to f3753f8 Compare April 9, 2024 09:42

flybird11111 closed this Apr 9, 2024

flybird11111 and others added 2 commits April 10, 2024 16:04

Merge branch 'feature/resize-embedding' into padding-vocab

1114509

[pre-commit.ci] auto fixes from pre-commit.com hooks

b570f1a

for more information, see https://pre-commit.ci

flybird11111 force-pushed the padding-vocab branch 4 times, most recently from cd9585b to 5464925 Compare April 10, 2024 13:56

fix

f08e084

fix fix fix fix fix fix fix

flybird11111 force-pushed the padding-vocab branch from ea8bcdc to f08e084 Compare April 10, 2024 15:20

ver217 reviewed Apr 11, 2024

View reviewed changes

flybird11111 force-pushed the padding-vocab branch 3 times, most recently from 16b6aa0 to 815d3cf Compare April 11, 2024 11:42

resolve comments

14a4342

resolve comments resolve comments resolve comments resolve comments

flybird11111 force-pushed the padding-vocab branch from 815d3cf to 14a4342 Compare April 12, 2024 04:01

ver217 reviewed Apr 12, 2024

View reviewed changes

colossalai/checkpoint_io/hybrid_parallel_checkpoint_io.py Outdated Show resolved Hide resolved

flybird11111 force-pushed the padding-vocab branch from 175fd26 to 935c2b3 Compare April 12, 2024 13:09

ver217 reviewed Apr 15, 2024

View reviewed changes

flybird11111 force-pushed the padding-vocab branch 4 times, most recently from 3b4f14a to d8c0a58 Compare April 16, 2024 09:47

ver217 reviewed Apr 17, 2024

View reviewed changes

tests/test_tensor/test_padded_tensor/test_padded_tensor.py Outdated Show resolved Hide resolved

tests/test_tensor/test_padded_tensor/test_padded_tensor.py Outdated Show resolved Hide resolved

tests/test_tensor/test_padded_tensor/test_padded_tensor.py Outdated Show resolved Hide resolved

flybird11111 force-pushed the padding-vocab branch 2 times, most recently from 0e6a979 to bdd917c Compare April 17, 2024 03:50

ptensor

873e2b3

ptensor resolve comments fix fix fix fix fix resolve comments resolve comments resolve comments resolve comments resolve comments

flybird11111 force-pushed the padding-vocab branch from bdd917c to 873e2b3 Compare April 17, 2024 06:30

ver217 approved these changes Apr 17, 2024

View reviewed changes

flybird11111 merged commit 6911f61 into hpcaitech:feature/resize-embedding Apr 17, 2024
3 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[shardformer]enable padding vocabulary size. #5489

[shardformer]enable padding vocabulary size. #5489

flybird11111 commented Mar 22, 2024

flybird11111 commented Mar 29, 2024

flybird11111 commented Apr 2, 2024 •

edited

Loading

flybird11111 commented Apr 3, 2024

insujang commented Apr 8, 2024

flybird11111 commented Apr 9, 2024

[shardformer]enable padding vocabulary size. #5489

[shardformer]enable padding vocabulary size. #5489

Conversation

flybird11111 commented Mar 22, 2024

📌 Checklist before creating the PR

🚨 Issue number

📝 What does this PR do?

💥 Checklist before requesting a review

⭐️ Do you enjoy contributing to Colossal-AI?

flybird11111 commented Mar 29, 2024

flybird11111 commented Apr 2, 2024 • edited Loading

flybird11111 commented Apr 3, 2024

insujang commented Apr 8, 2024

flybird11111 commented Apr 9, 2024

flybird11111 commented Apr 2, 2024 •

edited

Loading