biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-12 17:51:31 +01:00
committed by GitHub
parent 9bb38130cb
commit 8fcaaf6a16
944 changed files with 9490 additions and 10121 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
vllm/model_executor/models/lfm2_moe.py
View File

@@ -2,7 +2,7 @@
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from collections.abc import Iterable
from itertools import islice
from typing import Any, Optional
from typing import Any
import torch
import torch.nn as nn
@@ -65,7 +65,7 @@ class Lfm2MoeMlp(nn.Module):
self,
dim: int,
ff_dim: int,
quant_config: Optional[QuantizationConfig] = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",
):
super().__init__()
@@ -96,7 +96,7 @@ class Lfm2MoeSparseMoeBlock(nn.Module):
def __init__(
self,
config: Lfm2MoeConfig,
quant_config: Optional[QuantizationConfig] = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",
enable_eplb: bool = False,
):
@@ -191,10 +191,10 @@ class Lfm2MoeAttention(nn.Module):
num_heads: int,
num_kv_heads: int,
rope_theta: float = 10000,
rope_scaling: Optional[dict[str, Any]] = None,
rope_scaling: dict[str, Any] | None = None,
max_position_embeddings: int = 8192,
cache_config: Optional[CacheConfig] = None,
quant_config: Optional[QuantizationConfig] = None,
cache_config: CacheConfig | None = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",
) -> None:
super().__init__()
@@ -283,9 +283,9 @@ class Lfm2MoeAttentionDecoderLayer(nn.Module):
self,
config: Lfm2MoeConfig,
layer_idx: int,
model_config: Optional[ModelConfig] = None,
cache_config: Optional[CacheConfig] = None,
quant_config: Optional[QuantizationConfig] = None,
model_config: ModelConfig | None = None,
cache_config: CacheConfig | None = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",
enable_eplb: bool = False,
) -> None:
@@ -340,7 +340,7 @@ class Lfm2MoeAttentionDecoderLayer(nn.Module):
self,
positions: torch.Tensor,
hidden_states: torch.Tensor,
residual: Optional[torch.Tensor],
residual: torch.Tensor | None,
**kwargs,
) -> tuple[torch.Tensor, torch.Tensor]:
if residual is None:
@@ -358,9 +358,9 @@ class Lfm2MoeShortConvDecoderLayer(nn.Module):
self,
config: Lfm2MoeConfig,
layer_idx: int,
model_config: Optional[ModelConfig] = None,
cache_config: Optional[CacheConfig] = None,
quant_config: Optional[QuantizationConfig] = None,
model_config: ModelConfig | None = None,
cache_config: CacheConfig | None = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",
enable_eplb: bool = False,
) -> None:
@@ -396,7 +396,7 @@ class Lfm2MoeShortConvDecoderLayer(nn.Module):
def forward(
self,
hidden_states: torch.Tensor,
residual: Optional[torch.Tensor],
residual: torch.Tensor | None,
**kwargs,
):
if residual is None:
@@ -479,8 +479,8 @@ class Lfm2MoeModel(nn.Module):
self,
input_ids: torch.Tensor,
positions: torch.Tensor,
intermediate_tensors: Optional[IntermediateTensors] = None,
inputs_embeds: Optional[torch.Tensor] = None,
intermediate_tensors: IntermediateTensors | None = None,
inputs_embeds: torch.Tensor | None = None,
) -> torch.Tensor:
if get_pp_group().is_first_rank:
if inputs_embeds is not None:
@@ -774,8 +774,8 @@ class Lfm2MoeForCausalLM(
self,
input_ids: torch.Tensor,
positions: torch.Tensor,
intermediate_tensors: Optional[IntermediateTensors] = None,
inputs_embeds: Optional[torch.Tensor] = None,
intermediate_tensors: IntermediateTensors | None = None,
inputs_embeds: torch.Tensor | None = None,
**kwargs,
) -> torch.Tensor:
hidden_states = self.model(