ColossalAI/colossalai/shardformer/modeling
FoolPlayer dd2bf02679 [shardformer] support SAM (#4231)
* 1.support sam 2.add fused qkv for nn.Linear

* update utils support set element in list

* overtwrite SamVisionAttention foward to use DropoutForParallelInput

* remove unused code
2023-08-15 23:25:14 +08:00
..
__init__.py [shardformer] added development protocol for standardization (#4149) 2023-07-04 16:05:01 +08:00
bert.py [pipeline] fix return_dict/fix pure_pipeline_test (#4331) 2023-08-15 23:25:14 +08:00
bloom.py [pipeline] fix return_dict/fix pure_pipeline_test (#4331) 2023-08-15 23:25:14 +08:00
gpt2.py [pipeline] fix return_dict/fix pure_pipeline_test (#4331) 2023-08-15 23:25:14 +08:00
llama.py [plugin] add 3d parallel plugin (#4295) 2023-08-15 23:25:14 +08:00
sam.py [shardformer] support SAM (#4231) 2023-08-15 23:25:14 +08:00
t5.py [pipeline] add pipeline support for all T5 models (#4310) 2023-08-15 23:25:14 +08:00
vit.py [shardformer] support pipeline base vit model (#4284) 2023-08-15 23:25:14 +08:00