[MoE Refactor] Rename FusedMoE to FusedMoEFactory (#44941)
Signed-off-by: Bill Nell <bnell@redhat.com> Signed-off-by: bnellnm <49004751+bnellnm@users.noreply.github.com> Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
@@ -32,7 +32,9 @@ from vllm.distributed import (
|
||||
)
|
||||
from vllm.model_executor.layers.activation import SiluAndMul
|
||||
from vllm.model_executor.layers.attention import Attention
|
||||
from vllm.model_executor.layers.fused_moe import FusedMoE
|
||||
from vllm.model_executor.layers.fused_moe import (
|
||||
FusedMoEFactory,
|
||||
)
|
||||
from vllm.model_executor.layers.layernorm import RMSNorm
|
||||
from vllm.model_executor.layers.linear import (
|
||||
MergedColumnParallelLinear,
|
||||
@@ -325,7 +327,7 @@ class Param2MoEMoEBlock(nn.Module):
|
||||
else:
|
||||
self.shared_experts = None # type: ignore[assignment]
|
||||
|
||||
self.experts = FusedMoE(
|
||||
self.experts = FusedMoEFactory(
|
||||
shared_experts=self.shared_experts,
|
||||
num_experts=self.num_experts,
|
||||
top_k=self.top_k,
|
||||
@@ -342,7 +344,7 @@ class Param2MoEMoEBlock(nn.Module):
|
||||
routed_scaling_factor=self.routed_scaling_factor,
|
||||
)
|
||||
|
||||
def maybe_get_fused_moe(self) -> FusedMoE:
|
||||
def maybe_get_fused_moe(self) -> FusedMoEFactory:
|
||||
return self.experts
|
||||
|
||||
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
|
||||
|
||||
Reference in New Issue
Block a user