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:
+20
-22
@@ -8,7 +8,7 @@ import random
|
||||
import sys
|
||||
from datetime import timedelta
|
||||
from platform import uname
|
||||
from typing import TYPE_CHECKING, Any, NamedTuple, Optional, Union
|
||||
from typing import TYPE_CHECKING, Any, NamedTuple
|
||||
|
||||
import numpy as np
|
||||
import torch
|
||||
@@ -20,18 +20,16 @@ from vllm.logger import init_logger
|
||||
if TYPE_CHECKING:
|
||||
from vllm.attention.backends.registry import _Backend
|
||||
from vllm.config import ModelConfig, VllmConfig
|
||||
from vllm.lora.request import LoRARequest
|
||||
from vllm.pooling_params import PoolingParams
|
||||
from vllm.sampling_params import SamplingParams
|
||||
from vllm.utils import FlexibleArgumentParser
|
||||
else:
|
||||
_Backend = None
|
||||
ModelConfig = None
|
||||
VllmConfig = None
|
||||
LoRARequest = None
|
||||
PoolingParams = None
|
||||
SamplingParams = None
|
||||
FlexibleArgumentParser = None
|
||||
_Backend = object
|
||||
ModelConfig = object
|
||||
VllmConfig = object
|
||||
PoolingParams = object
|
||||
SamplingParams = object
|
||||
FlexibleArgumentParser = object
|
||||
|
||||
logger = init_logger(__name__)
|
||||
|
||||
@@ -113,7 +111,7 @@ class Platform:
|
||||
|
||||
additional_env_vars: list[str] = []
|
||||
|
||||
_global_graph_pool: Optional[Any] = None
|
||||
_global_graph_pool: Any | None = None
|
||||
|
||||
@property
|
||||
def supported_dtypes(self) -> list[torch.dtype]:
|
||||
@@ -180,7 +178,7 @@ class Platform:
|
||||
import vllm._moe_C # noqa: F401
|
||||
|
||||
@classmethod
|
||||
def get_vit_attn_backend(cls, head_size: int, dtype: torch.dtype) -> "_Backend":
|
||||
def get_vit_attn_backend(cls, head_size: int, dtype: torch.dtype) -> _Backend:
|
||||
from vllm.attention.backends.registry import _Backend
|
||||
|
||||
return _Backend.TORCH_SDPA
|
||||
@@ -188,10 +186,10 @@ class Platform:
|
||||
@classmethod
|
||||
def get_attn_backend_cls(
|
||||
cls,
|
||||
selected_backend: "_Backend",
|
||||
selected_backend: _Backend,
|
||||
head_size: int,
|
||||
dtype: torch.dtype,
|
||||
kv_cache_dtype: Optional[str],
|
||||
kv_cache_dtype: str | None,
|
||||
block_size: int,
|
||||
use_v1: bool,
|
||||
use_mla: bool,
|
||||
@@ -205,14 +203,14 @@ class Platform:
|
||||
def get_device_capability(
|
||||
cls,
|
||||
device_id: int = 0,
|
||||
) -> Optional[DeviceCapability]:
|
||||
) -> DeviceCapability | None:
|
||||
"""Stateless version of [torch.cuda.get_device_capability][]."""
|
||||
return None
|
||||
|
||||
@classmethod
|
||||
def has_device_capability(
|
||||
cls,
|
||||
capability: Union[tuple[int, int], int],
|
||||
capability: tuple[int, int] | int,
|
||||
device_id: int = 0,
|
||||
) -> bool:
|
||||
"""
|
||||
@@ -236,7 +234,7 @@ class Platform:
|
||||
@classmethod
|
||||
def is_device_capability(
|
||||
cls,
|
||||
capability: Union[tuple[int, int], int],
|
||||
capability: tuple[int, int] | int,
|
||||
device_id: int = 0,
|
||||
) -> bool:
|
||||
"""
|
||||
@@ -283,7 +281,7 @@ class Platform:
|
||||
return torch.inference_mode(mode=True)
|
||||
|
||||
@classmethod
|
||||
def seed_everything(cls, seed: Optional[int] = None) -> None:
|
||||
def seed_everything(cls, seed: int | None = None) -> None:
|
||||
"""
|
||||
Set the seed of each random module.
|
||||
`torch.manual_seed` will set seed on all devices.
|
||||
@@ -304,7 +302,7 @@ class Platform:
|
||||
|
||||
@classmethod
|
||||
def pre_register_and_update(
|
||||
cls, parser: Optional[FlexibleArgumentParser] = None
|
||||
cls, parser: FlexibleArgumentParser | None = None
|
||||
) -> None:
|
||||
"""
|
||||
Do some pre-registration or update action for the current platform.
|
||||
@@ -389,7 +387,7 @@ class Platform:
|
||||
|
||||
@classmethod
|
||||
def get_current_memory_usage(
|
||||
cls, device: Optional[torch.types.Device] = None
|
||||
cls, device: torch.types.Device | None = None
|
||||
) -> float:
|
||||
"""
|
||||
Return the memory usage in bytes.
|
||||
@@ -501,7 +499,7 @@ class Platform:
|
||||
def validate_request(
|
||||
cls,
|
||||
prompt: PromptType,
|
||||
params: Union[SamplingParams, PoolingParams],
|
||||
params: SamplingParams | PoolingParams,
|
||||
processed_inputs: ProcessorInputs,
|
||||
) -> None:
|
||||
"""Raises if this request is unsupported on this platform"""
|
||||
@@ -557,7 +555,7 @@ class Platform:
|
||||
|
||||
@classmethod
|
||||
def is_kv_cache_dtype_supported(
|
||||
cls, kv_cache_dtype: str, model_config: "ModelConfig"
|
||||
cls, kv_cache_dtype: str, model_config: ModelConfig
|
||||
) -> bool:
|
||||
"""
|
||||
Returns if the kv_cache_dtype is supported by the current platform.
|
||||
@@ -617,7 +615,7 @@ class Platform:
|
||||
return {}
|
||||
|
||||
@classmethod
|
||||
def get_nixl_memory_type(cls) -> Optional[str]:
|
||||
def get_nixl_memory_type(cls) -> str | None:
|
||||
"""
|
||||
Returns the nixl memory type for the current platform.
|
||||
"""
|
||||
|
||||
Reference in New Issue
Block a user