FoolPlayer 92f6791095 [shardformer] Add layernorm (#4072) ... * 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		2023-07-04 16:05:01 +08:00
..
__init__.py	[shardformer] Add layernorm (#4072)	2023-07-04 16:05:01 +08:00
_operation.py	[shardformer] supported fused qkv checkpoint (#4073)	2023-07-04 16:05:01 +08:00
dropout.py	[shardformer] refactored the shardformer layer structure (#4053)	2023-07-04 16:05:01 +08:00
embedding.py	[shardformer] supported fused qkv checkpoint (#4073)	2023-07-04 16:05:01 +08:00
layernorm.py	[shardformer] Add layernorm (#4072)	2023-07-04 16:05:01 +08:00
linear.py	[shardformer] supported fused qkv checkpoint (#4073)	2023-07-04 16:05:01 +08:00
linear_conv.py	[shardformer] supported fused qkv checkpoint (#4073)	2023-07-04 16:05:01 +08:00
loss.py	[shardformer] refactored the shardformer layer structure (#4053)	2023-07-04 16:05:01 +08:00
parallel_module.py	[shardformer] supported fused qkv checkpoint (#4073)	2023-07-04 16:05:01 +08:00
utils.py	[shardformer] integrated linear 1D with dtensor (#3996)	2023-07-04 16:05:01 +08:00