[Attention][V0 Deprecation] Deprecate accept output buffer (#39125)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
This commit is contained in:
@@ -94,7 +94,6 @@ def basic_cache(
|
||||
class CacheOnlyAttentionBackend(AttentionBackend):
|
||||
"""Attention backend that only caches KV without computing attention."""
|
||||
|
||||
accept_output_buffer: bool = False
|
||||
supported_dtypes: ClassVar[list[torch.dtype]] = [
|
||||
torch.float16,
|
||||
torch.bfloat16,
|
||||
|
||||
@@ -184,7 +184,7 @@ def create_whisper_attention_backend_with_block_pooling(
|
||||
value: torch.Tensor,
|
||||
kv_cache: torch.Tensor,
|
||||
attn_metadata: AttentionMetadata,
|
||||
output: torch.Tensor | None = None,
|
||||
output: torch.Tensor,
|
||||
output_scale: torch.Tensor | None = None,
|
||||
output_block_scale: torch.Tensor | None = None,
|
||||
) -> torch.Tensor:
|
||||
|
||||
Reference in New Issue
Block a user