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

Update deprecated type hinting in models (#18132)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-05-15 06:06:50 +01:00
committed by GitHub
parent 83f74c698f
commit 26d0419309
130 changed files with 971 additions and 901 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
vllm/model_executor/models/qwen2_5_vl.py
View File

@@ -24,9 +24,9 @@
# See the License for the specific language governing permissions and
# limitations under the License.
"""Inference-only Qwen2.5-VL model compatible with HuggingFace weights."""
from collections.abc import Iterable, Mapping
from functools import partial
from typing import (Callable, Iterable, List, Literal, Mapping, Optional, Set,
Tuple, TypedDict, Union)
from typing import Callable, Literal, Optional, TypedDict, Union
import torch
import torch.nn as nn
@@ -91,7 +91,7 @@ class Qwen2_5_VLImageEmbeddingInputs(TypedDict):
type: Literal["image_embeds"]
image_embeds: torch.Tensor
"""Supported types:
- List[`torch.Tensor`]: A list of tensors holding all images' features.
- list[`torch.Tensor`]: A list of tensors holding all images' features.
Each tensor holds an image's features.
- `torch.Tensor`: A tensor holding all images' features
(concatenation of all images' feature tensors).
@@ -137,7 +137,7 @@ class Qwen2_5_VLVideoEmbeddingInputs(TypedDict):
type: Literal["video_embeds"]
video_embeds: torch.Tensor
"""Supported types:
- List[`torch.Tensor`]: A list of tensors holding all videos' features.
- list[`torch.Tensor`]: A list of tensors holding all videos' features.
Each tensor holds an video's features.
- `torch.Tensor`: A tensor holding all videos' features
(concatenation of all videos' feature tensors).
@@ -709,8 +709,8 @@ class Qwen2_5_VisionTransformer(nn.Module):
hidden_states = hidden_states[reverse_indices, :]
return hidden_states
def load_weights(self, weights: Iterable[Tuple[str,
torch.Tensor]]) -> Set[str]:
def load_weights(self, weights: Iterable[tuple[str,
torch.Tensor]]) -> set[str]:
stacked_params_mapping = [
# (param_name, shard_name, shard_id)
("attn.qkv.", "attn.q.", "q"),
@@ -718,7 +718,7 @@ class Qwen2_5_VisionTransformer(nn.Module):
("attn.qkv.", "attn.v.", "v"),
]
params_dict = dict(self.named_parameters(remove_duplicate=False))
loaded_params: Set[str] = set()
loaded_params: set[str] = set()
for name, loaded_weight in weights:
for (param_name, weight_name, shard_id) in stacked_params_mapping:
@@ -750,7 +750,7 @@ class Qwen2_5_VLProcessingInfo(Qwen2VLProcessingInfo):
min_pixels: Optional[int] = None,
max_pixels: Optional[int] = None,
size: Optional[dict[str, int]] = None,
fps: Optional[Union[float, List[float]]] = None,
fps: Optional[Union[float, list[float]]] = None,
**kwargs: object,
) -> Qwen2_5_VLProcessor:
if fps is not None:
@@ -1116,8 +1116,8 @@ class Qwen2_5_VLForConditionalGeneration(nn.Module, SupportsMultiModal,
return self.language_model.compute_logits(hidden_states,
sampling_metadata)
def load_weights(self, weights: Iterable[Tuple[str,
torch.Tensor]]) -> Set[str]:
def load_weights(self, weights: Iterable[tuple[str,
torch.Tensor]]) -> set[str]:
loader = AutoWeightsLoader(self)
return loader.load_weights(weights, mapper=self.hf_to_vllm_mapper)