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

33
vllm/model_executor/models/granitemoehybrid.py
View File

@@ -4,7 +4,6 @@
# Added by the IBM Team, 2025
from collections.abc import Iterable
from typing import Optional
import torch
from torch import nn
@@ -50,9 +49,9 @@ class GraniteMoeHybridMambaDecoderLayer(nn.Module):
self,
config: GraniteMoeHybridConfig,
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 = "",
) -> None:
super().__init__()
@@ -105,7 +104,7 @@ class GraniteMoeHybridMambaDecoderLayer(nn.Module):
def forward(
self,
hidden_states: torch.Tensor,
residual: Optional[torch.Tensor],
residual: torch.Tensor | None,
**kwargs,
):
residual = hidden_states
@@ -139,9 +138,9 @@ class GraniteMoeHybridAttentionDecoderLayer(nn.Module):
self,
config: GraniteMoeHybridConfig,
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 = "",
) -> None:
super().__init__()
@@ -183,7 +182,7 @@ class GraniteMoeHybridAttentionDecoderLayer(nn.Module):
self,
positions: torch.Tensor,
hidden_states: torch.Tensor,
residual: Optional[torch.Tensor],
residual: torch.Tensor | None,
) -> torch.Tensor:
residual = hidden_states
hidden_states = self.input_layernorm(hidden_states)
@@ -218,9 +217,9 @@ class GraniteMoeHybridAttention(nn.Module):
def __init__(
self,
config: GraniteMoeHybridConfig,
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 = "",
) -> None:
super().__init__()
@@ -374,8 +373,8 @@ class GraniteMoeHybridModel(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:
@@ -614,8 +613,8 @@ class GraniteMoeHybridForCausalLM(
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,
):
hidden_states = self.model(
@@ -627,7 +626,7 @@ class GraniteMoeHybridForCausalLM(
def compute_logits(
self,
hidden_states: torch.Tensor,
) -> Optional[torch.Tensor]:
) -> torch.Tensor | None:
logits = self.logits_processor(self.lm_head, hidden_states)
return logits