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

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-12 09:51:31 -07:00
committed by GitHub
parent 9bb38130cb
commit 8fcaaf6a16
944 changed files with 9490 additions and 10121 deletions
+18 -20
View File
@@ -26,7 +26,7 @@
from collections.abc import Iterable
from itertools import islice
from typing import Any, Optional, Union
from typing import Any
import torch
from torch import nn
@@ -76,7 +76,7 @@ class LlamaMLP(nn.Module):
hidden_size: int,
intermediate_size: int,
hidden_act: str,
quant_config: Optional[QuantizationConfig] = None,
quant_config: QuantizationConfig | None = None,
bias: bool = False,
prefix: str = "",
reduce_results: bool = True,
@@ -121,12 +121,12 @@ class LlamaAttention(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,
quant_config: Optional[QuantizationConfig] = None,
quant_config: QuantizationConfig | None = None,
bias: bool = False,
bias_o_proj: bool = False,
cache_config: Optional[CacheConfig] = None,
cache_config: CacheConfig | None = None,
prefix: str = "",
attn_type: str = AttentionType.DECODER,
) -> None:
@@ -236,8 +236,8 @@ class LlamaAttention(nn.Module):
def _init_rotary_emb(
self,
config: LlamaConfig,
rope_scaling: Optional[dict[str, Any]],
quant_config: Optional[QuantizationConfig],
rope_scaling: dict[str, Any] | None,
quant_config: QuantizationConfig | None,
) -> None:
is_neox_style = True
is_gguf = quant_config and quant_config.get_name() == "gguf"
@@ -260,7 +260,7 @@ class LlamaDecoderLayer(nn.Module):
self,
vllm_config: VllmConfig,
prefix: str = "",
config: Optional[LlamaConfig] = None,
config: LlamaConfig | None = None,
) -> None:
super().__init__()
@@ -331,7 +331,7 @@ class LlamaDecoderLayer(nn.Module):
self,
positions: torch.Tensor,
hidden_states: torch.Tensor,
residual: Optional[torch.Tensor],
residual: torch.Tensor | None,
) -> tuple[torch.Tensor, torch.Tensor]:
# Self Attention
if residual is None:
@@ -346,7 +346,7 @@ class LlamaDecoderLayer(nn.Module):
hidden_states = self.mlp(hidden_states)
return hidden_states, residual
def get_quant_config(self, vllm_config: VllmConfig) -> Optional[QuantizationConfig]:
def get_quant_config(self, vllm_config: VllmConfig) -> QuantizationConfig | None:
"""Get quantization config for this layer. Override in subclasses."""
return vllm_config.quant_config
@@ -407,13 +407,11 @@ class LlamaModel(nn.Module):
def forward(
self,
input_ids: Optional[torch.Tensor],
input_ids: torch.Tensor | None,
positions: torch.Tensor,
intermediate_tensors: Optional[IntermediateTensors],
inputs_embeds: Optional[torch.Tensor] = None,
) -> Union[
torch.Tensor, IntermediateTensors, tuple[torch.Tensor, list[torch.Tensor]]
]:
intermediate_tensors: IntermediateTensors | None,
inputs_embeds: torch.Tensor | None = None,
) -> torch.Tensor | IntermediateTensors | tuple[torch.Tensor, list[torch.Tensor]]:
if get_pp_group().is_first_rank:
if inputs_embeds is not None:
hidden_states = inputs_embeds
@@ -627,9 +625,9 @@ class LlamaForCausalLM(nn.Module, SupportsLoRA, SupportsPP, SupportsEagle3):
self,
input_ids: torch.Tensor,
positions: torch.Tensor,
intermediate_tensors: Optional[IntermediateTensors] = None,
inputs_embeds: Optional[torch.Tensor] = None,
) -> Union[torch.Tensor, IntermediateTensors]:
intermediate_tensors: IntermediateTensors | None = None,
inputs_embeds: torch.Tensor | None = None,
) -> torch.Tensor | IntermediateTensors:
model_output = self.model(
input_ids, positions, intermediate_tensors, inputs_embeds
)
@@ -638,7 +636,7 @@ class LlamaForCausalLM(nn.Module, SupportsLoRA, SupportsPP, SupportsEagle3):
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