Commit Graph

133 Commits (b5f9e37c709656b286940f1b5e05abddfa257e3d)

Author SHA1 Message Date
Frank Lee c1d5453e9f [shardformer] adapted llama to the new API (#4036)
1 year ago
FoolPlayer 74d176c8d8 [shardformer] fix bert and gpt downstream with new api (#4024)
1 year ago
Frank Lee e253a07007 [shardformer] updated doc (#4016)
1 year ago
FoolPlayer df018fc305 support bert with new api
1 year ago
FoolPlayer 507c0ad368 add vocabembedding layer
1 year ago
Frank Lee 45d9384346 [shardformer] removed inplace tensor sharding (#4018)
1 year ago
Frank Lee 3893fa1a8d [shardformer] refactored embedding and dropout to parallel module (#4013)
1 year ago
FoolPlayer dfca9678fa integrate with dist layer (#4011)
1 year ago
Frank Lee 015af592f8 [shardformer] integrated linear 1D with dtensor (#3996)
1 year ago
FoolPlayer d3bc530849 [shardformer] Refactor shardformer api (#4001)
1 year ago
FoolPlayer f7774ec0f3 [Shardformer] Downstream bert (#3979)
1 year ago
wukong1992 c1c672d0f0 [shardformer] shardformer support t5 model (#3994)
1 year ago
wukong1992 6b30dfb7ce [shardformer] support llama model using shardformer (#3969)
1 year ago
FoolPlayer 45927d5527 [shardformer] Add dropout layer in shard model and refactor policy api (#3949)
1 year ago
FoolPlayer a73130482d [shardformer] Unit test (#3928)
1 year ago
FoolPlayer f1cb5ac6bf [shardformer] Align bert value (#3907)
1 year ago
FoolPlayer 79f8d5d54b [shardformer] add gpt2 policy and modify shard and slicer to support (#3883)
1 year ago
FoolPlayer 70173e3123 update README (#3909)
1 year ago
FoolPlayer ab8a47f830 [shardformer] add Dropout layer support different dropout pattern (#3856)
1 year ago
FoolPlayer c594dc2f1c [shardformer] update readme with modules implement doc (#3834)
1 year ago
Frank Lee 4972e1f40e [shardformer] refactored the user api (#3828)
1 year ago
Frank Lee 235792f170 [shardformer] updated readme (#3827)
1 year ago
FoolPlayer 8cc11235c0 [shardformer]: Feature/shardformer, add some docstring and readme (#3816)
1 year ago
FoolPlayer 8d68de767d [shardformer] init shardformer code structure (#3731)
1 year ago
Frank Lee ddcf58cacf
Revert "[sync] sync feature/shardformer with develop"
1 year ago
FoolPlayer ef1537759c [shardformer] add gpt2 policy and modify shard and slicer to support (#3883)
1 year ago
FoolPlayer 6370a935f6 update README (#3909)
1 year ago
FoolPlayer 21a3915c98 [shardformer] add Dropout layer support different dropout pattern (#3856)
1 year ago
FoolPlayer 997544c1f9 [shardformer] update readme with modules implement doc (#3834)
1 year ago
Frank Lee 537a52b7a2 [shardformer] refactored the user api (#3828)
1 year ago
Frank Lee bc19024bf9 [shardformer] updated readme (#3827)
1 year ago
FoolPlayer 58f6432416 [shardformer]: Feature/shardformer, add some docstring and readme (#3816)
1 year ago
FoolPlayer 6a69b44dfc [shardformer] init shardformer code structure (#3731)
1 year ago