Commit Graph

9 Commits (451e9142b8b8b77ed3138fb03ad54494c3c57126)

Author SHA1 Message Date
Elsa Granger b2ad0d9e8f
[pipeline,shardformer] Fix p2p efficiency in pipeline, allow skipping loading weight not in weight_map when `strict=False`, fix llama flash attention forward, add flop estimation by megatron in llama benchmark (#5017)
* Use p2p

* Cannot bidirectonal send p2p

* Refactor tensor creation and serialization in P2P
communication

* Fix llama forward args in flash attention

* Add flop estimate from megatron

* Support loading weight not in weight_map when strict=False in hybrid_parallel

* Use send_forward_recv_backward, etc in 1f1b

* Use dataclass for metdata
Remove torch.cuda.synchronize() as suggested

* Add comment about the torch.cuda.synchronize for potential error

* Typo

* Update hybrid_parallel_checkpoint_io.py

* Update p2p.py

* Update one_f_one_b.py

* Update p2p.py

---------

Co-authored-by: flybird11111 <1829166702@qq.com>
2023-11-16 20:15:59 +08:00
Baizhou Zhang 64a08b2dc3
[checkpointio] support unsharded checkpointIO for hybrid parallel (#4774)
* support unsharded saving/loading for model

* support optimizer unsharded saving

* update doc

* support unsharded loading for optimizer

* small fix
2023-09-26 10:58:03 +08:00
Baizhou Zhang c0a033700c
[shardformer] fix master param sync for hybrid plugin/rewrite unwrapping logic (#4758)
* fix master param sync for hybrid plugin

* rewrite unwrap for ddp/fsdp

* rewrite unwrap for zero/gemini

* rewrite unwrap for hybrid plugin

* fix geemini unwrap

* fix bugs
2023-09-20 18:29:37 +08:00
Hongxin Liu 079bf3cb26
[misc] update pre-commit and run all files (#4752)
* [misc] update pre-commit

* [misc] run pre-commit

* [misc] remove useless configuration files

* [misc] ignore cuda for clang-format
2023-09-19 14:20:26 +08:00
flybird11111 4c4482f3ad
[example] llama2 add fine-tune example (#4673)
* [shardformer] update shardformer readme

[shardformer] update shardformer readme

[shardformer] update shardformer readme

* [shardformer] update llama2/opt finetune example and shardformer update to llama2

* [shardformer] update llama2/opt finetune example and shardformer update to llama2

* [shardformer] update llama2/opt finetune example and shardformer update to llama2

* [shardformer] change dataset

* [shardformer] change dataset

* [shardformer] fix CI

* [shardformer] fix

* [shardformer] fix

* [shardformer] fix

* [shardformer] fix

* [shardformer] fix

[example] update opt example

[example] resolve comments

fix

fix

* [example] llama2 add finetune example

* [example] llama2 add finetune example

* [example] llama2 add finetune example

* [example] llama2 add finetune example

* fix

* update llama2 example

* update llama2 example

* fix

* update llama2 example

* update llama2 example

* update llama2 example

* update llama2 example

* update llama2 example

* update llama2 example

* Update requirements.txt

* update llama2 example

* update llama2 example

* update llama2 example
2023-09-15 18:45:44 +08:00
Baizhou Zhang d8ceeac14e
[hotfix] fix typo in hybrid parallel io (#4697) 2023-09-12 17:32:19 +08:00
Baizhou Zhang 38ccb8b1a3
[shardformer] support from_pretrained when loading model with HybridParallelPlugin (#4575)
* hybrid plugin support huggingface from_pretrained

* add huggingface compatibility tests

* add folder cleaning

* fix bugs
2023-09-01 17:40:01 +08:00
Baizhou Zhang c9625dbb63
[shardformer] support sharded optimizer checkpointIO of HybridParallelPlugin (#4540)
* implement sharded optimizer saving

* add more param info

* finish implementation of sharded optimizer saving

* fix bugs in optimizer sharded saving

* add pp+zero test

* param group loading

* greedy loading of optimizer

* fix bug when loading

* implement optimizer sharded saving

* add optimizer test & arrange checkpointIO utils

* fix gemini sharding state_dict

* add verbose option

* add loading of master params

* fix typehint

* fix master/working mapping in fp16 amp
2023-08-31 14:50:47 +08:00
Baizhou Zhang 44eab2b27f
[shardformer] support sharded checkpoint IO for models of HybridParallelPlugin (#4506)
* add APIs

* implement save_sharded_model

* add test for hybrid checkpointio

* implement naive loading for sharded model

* implement efficient sharded model loading

* open a new file for hybrid checkpoint_io

* small fix

* fix circular importing

* fix docstring

* arrange arguments and apis

* small fix
2023-08-25 22:04:57 +08:00