[5/N][Attention] Finish eliminating vllm/attention folder (#32064)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
This commit is contained in:
@@ -21,7 +21,6 @@ from tests.compile.fusion_test_utils import (
|
||||
from tests.utils import flat_product
|
||||
from tests.v1.attention.utils import BatchSpec, create_common_attn_metadata
|
||||
from vllm._custom_ops import cutlass_scaled_fp4_mm, scaled_fp4_quant
|
||||
from vllm.attention.layer import Attention
|
||||
from vllm.compilation.fusion_attn import ATTN_OP, AttnFusionPass
|
||||
from vllm.compilation.fx_utils import find_op_nodes
|
||||
from vllm.compilation.matcher_utils import QUANT_OPS
|
||||
@@ -40,6 +39,7 @@ from vllm.config import (
|
||||
set_current_vllm_config,
|
||||
)
|
||||
from vllm.forward_context import get_forward_context, set_forward_context
|
||||
from vllm.model_executor.layers.attention import Attention
|
||||
from vllm.model_executor.layers.quantization.utils.quant_utils import (
|
||||
QuantKey,
|
||||
kFp8StaticTensorSym,
|
||||
|
||||
@@ -5,7 +5,6 @@ import pytest
|
||||
import torch
|
||||
|
||||
from tests.compile.backend import TestBackend
|
||||
from vllm.attention.layer import Attention
|
||||
from vllm.compilation.matcher_utils import FLASHINFER_ROTARY_OP, RMS_OP, ROTARY_OP
|
||||
from vllm.compilation.noop_elimination import NoOpEliminationPass
|
||||
from vllm.compilation.post_cleanup import PostCleanupPass
|
||||
@@ -21,6 +20,7 @@ from vllm.config import (
|
||||
VllmConfig,
|
||||
set_current_vllm_config,
|
||||
)
|
||||
from vllm.model_executor.layers.attention import Attention
|
||||
from vllm.model_executor.layers.layernorm import RMSNorm
|
||||
from vllm.model_executor.layers.rotary_embedding import RotaryEmbedding
|
||||
from vllm.platforms import current_platform
|
||||
|
||||
Reference in New Issue
Block a user