ColossalAI/colossalai/kernel/cuda_native/__init__.py
flybird1111 7a3dfd0c64 [shardformer] update shardformer to use flash attention 2 (#4392)
* cherry-pick flash attention 2

cherry-pick flash attention 2

* [shardformer] update shardformer to use flash attention 2

[shardformer] update shardformer to use flash attention 2, fix

[shardformer] update shardformer to use flash attention 2, fix

[shardformer] update shardformer to use flash attention 2, fix
2023-08-15 23:25:14 +08:00

10 lines
389 B
Python

from .layer_norm import MixedFusedLayerNorm as LayerNorm
from .mha.mha import ColoAttention
from .multihead_attention import MultiHeadAttention
from .scaled_softmax import AttnMaskType, FusedScaleMaskSoftmax, ScaledUpperTriangMaskedSoftmax
__all__ = [
'LayerNorm', 'MultiHeadAttention', 'FusedScaleMaskSoftmax', 'ScaledUpperTriangMaskedSoftmax', 'ColoAttention',
'AttnMaskType'
]