yuehuayingxueluo 249644c23b [Inference]Repalce Attention layer and MLP layer by shardformer to optimize the weight transpose operation，add fused_qkv and fused linear_add (#5340) ...

* add fused qkv

* replace attn and mlp by shardformer

* fix bugs in mlp

* add docstrings

* fix test_inference_engine.py

* add optimize unbind

* add fused_addmm

* rm squeeze(1)

* refactor codes

* fix ci bugs

* rename ShardFormerLlamaMLP and ShardFormerLlamaAttention

* Removed the dependency on LlamaFlashAttention2

* rollback test_inference_engine.py

2024-02-01 15:49:39 +08:00

..

jit

[npu] change device to accelerator api (#5239)

2024-01-09 10:20:05 +08:00

triton

[Inference]Repalce Attention layer and MLP layer by shardformer to optimize the weight transpose operation，add fused_qkv and fused linear_add (#5340)

2024-02-01 15:49:39 +08:00

__init__.py

[feat] refactored extension module (#5298)

2024-01-25 17:01:48 +08:00

extensions

[feat] refactored extension module (#5298)

2024-01-25 17:01:48 +08:00

kernel_loader.py

[feat] refactored extension module (#5298)

2024-01-25 17:01:48 +08:00