Wenwen Qu 21624f6f81 fix(moe): remove norm&gate force sync (#448) ... * add zero broadcast_sync * delete old sync logic * fix merged error * refactor code * remove some unused function (is norm/gate group)		2023-11-01 11:29:55 +08:00
..
__init__.py	feat(train): add fsdp training option (#293)	2023-10-09 18:59:31 +08:00
base_optimizer.py	feat(train): add fsdp training option (#293)	2023-10-09 18:59:31 +08:00
fsdp_optimizer.py	fix(optimizer/fsdp_optimizer.py): fsdp process empty params group (#408)	2023-10-10 20:06:04 +08:00
hybrid_zero_optim.py	fix(moe): remove norm&gate force sync (#448)	2023-11-01 11:29:55 +08:00
store.py	feat(moe):support zero for expert local dp (#404)	2023-10-09 17:45:26 +08:00
utils.py	fix(moe): remove norm&gate force sync (#448)	2023-11-01 11:29:55 +08:00