GitLab

* support shard optimizer of zero

* polish code

* support sync grad manually

* add state dict for zero

* fix unit test

* polish

* allow passing process group to zero12

* union tp-zero and normal-zero

* polish code

* support no sync for zero1 plugin

* polish

* polish

* refactor low level zero

* fix zero2 and support cpu offload

* avg gradient and modify unit test

* refactor grad store, support layer drop

* refactor bucket store, support grad accumulation

* fix and update unit test of zero and ddp

* compatible with tp, ga and unit test

* fix memory leak and polish

* add zero layer drop unittest

* polish code

* fix import err in unit test

* support diffenert comm dtype, modify docstring style

* polish code

* test padding and fix

* fix unit test of low level zero

* fix pad recording in bucket store

* support some models

* polish

* sharded optimizer checkpoint for gemini plugin

* modify test to reduce testing time

* update doc

* fix bug when keep_gatherd is true under GeminiPlugin

* [lazy] support init on cuda

* [test] update lazy init test

* [test] fix transformer version

* added softmax kernel

* added qkv_kernel

* added ops

* adding tests

* upload tets

* fix tests

* debugging

* debugging tests

* debugging

* added

* fixed errors

* added softmax kernel

* clean codes

* added tests

* update tests

* update tests

* added attention

* add

* fixed pytest checking

* add cuda check

* fix cuda version

* fix typo

* [checkpointio] unsharded optimizer checkpoint for Gemini plugin

* [checkpointio] unsharded optimizer checkpoint for Gemini using all_gather

Co-authored-by: github-actions <github-actions@github.com>

* [shardformer] made tensor parallelism configurable

* polish code

* [shardformer] refactored some doc and api

* polish code

* first v of vit shardformer

* keep vit

* update

* vit shard add vitattention vitlayer

* update num head shard para

* finish test for vit

* add new_model_class & postprocess

* add vit readme

* delete old files & fix the conflict

* fix sth

* [shardformer] shardformer support opt models

* [shardformer] shardformer support opt models, fix

* [shardformer] shardformer support opt models, fix

* [shardformer] shardformer support opt models, fix

* [test] fixed tests failed due to dtensor change

* polish code

* add layernorm to bert

* add layernorm test

* add layernorm test with load state dict

* add use_mixedfusedLN in shard config

* refactor policy to support fused_layernorm

* add linearconv1d test

* add linearconv1d test

* [shardformer] support module saving and loading

* polish code

* support kit use for bert test

* support kit test for gpt2

* [shardformer] adapted T5 and LLaMa test to use kit

* polish code

* add gpt2 test and layer class refactor

* add dropout in gpt2 policy

* fix bert downstream with new api

* remove comment line

* [shardformer] refactored embedding and dropout to parallel module

* polish code

* [shardformer] integrated linear 1D with dtensor

* polish code

* add dist dropout in model

* update docstring and bert policy with dropout

* refactor basepolicy and sharded, update bert

* update format

* update gpt2 policy

* update bert policy

* remove unused code

* update readme for new policy usage

* add downstream model of bert

* remove unused code

test t5

adjust layer attr

* fix bug in slicer, add slicer unit test

* add dropout test

* use pid as dropout seed

* updata dropout test with local pattern

* ad todo