[MoE][Refactor] Remove most arguments to FusedMoEMethodBase.apply (#29066)
Signed-off-by: Bill Nell <bnell@redhat.com> Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com> Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com> Co-authored-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
This commit is contained in:
co-authored by
Robert Shaw
Lucas Wilkinson
Tyler Michael Smith
parent
7618dc973d
commit
00e5cbb967
@@ -1,7 +1,7 @@
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
||||
|
||||
from collections.abc import Callable, Mapping
|
||||
from collections.abc import Mapping
|
||||
from types import MappingProxyType
|
||||
from typing import Any, Optional
|
||||
|
||||
@@ -625,26 +625,9 @@ class GGUFMoEMethod(FusedMoEMethodBase):
|
||||
layer: FusedMoE,
|
||||
x: torch.Tensor,
|
||||
router_logits: torch.Tensor,
|
||||
top_k: int,
|
||||
renormalize: bool,
|
||||
use_grouped_topk: bool = False,
|
||||
topk_group: int | None = None,
|
||||
num_expert_group: int | None = None,
|
||||
global_num_experts: int = -1,
|
||||
expert_map: torch.Tensor | None = None,
|
||||
custom_routing_function: Callable | None = None,
|
||||
scoring_func: str = "softmax",
|
||||
routed_scaling_factor: float = 1.0,
|
||||
e_score_correction_bias: torch.Tensor | None = None,
|
||||
apply_router_weight_on_input: bool = False,
|
||||
activation: str = "silu",
|
||||
enable_eplb: bool = False,
|
||||
expert_load_view: torch.Tensor | None = None,
|
||||
logical_to_physical_map: torch.Tensor | None = None,
|
||||
logical_replica_count: torch.Tensor | None = None,
|
||||
) -> torch.Tensor | tuple[torch.Tensor, torch.Tensor]:
|
||||
assert activation == "silu", "Only SiLU activation is supported."
|
||||
if apply_router_weight_on_input:
|
||||
assert layer.activation == "silu", "Only SiLU activation is supported."
|
||||
if layer.apply_router_weight_on_input:
|
||||
raise NotImplementedError(
|
||||
"Apply router weight on input is not supported for"
|
||||
"fused GGUF MoE method."
|
||||
@@ -662,7 +645,7 @@ class GGUFMoEMethod(FusedMoEMethodBase):
|
||||
topk_ids,
|
||||
layer.w13_qweight_type.weight_type,
|
||||
layer.w2_qweight_type.weight_type,
|
||||
activation,
|
||||
layer.activation,
|
||||
)
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user