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

13
vllm/model_executor/models/bert.py
View File

@@ -1,6 +1,7 @@
# SPDX-License-Identifier: Apache-2.0
from typing import Iterable, Optional, Set, Tuple
from collections.abc import Iterable
from typing import Optional
import torch
from torch import nn
@@ -349,8 +350,8 @@ class BertModel(nn.Module, SupportsQuant):
token_type_ids=token_type_ids)
return self.encoder(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)
("qkv_proj", "query", "q"),
@@ -359,7 +360,7 @@ class BertModel(nn.Module, SupportsQuant):
]
params_dict = dict(self.named_parameters())
loaded_params: Set[str] = set()
loaded_params: set[str] = set()
for name, loaded_weight in weights:
if self.pooler is None and "pooler" in name:
continue
@@ -424,7 +425,7 @@ class BertEmbeddingModel(nn.Module, SupportsV0Only, SupportsQuant):
) -> Optional[PoolerOutput]:
return self._pooler(hidden_states, pooling_metadata)
def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
weights = self.hf_to_vllm_mapper.apply(weights)
weights = ((name, data) for name, data in weights
if not name.startswith("lm_head."))
@@ -472,7 +473,7 @@ class BertForSequenceClassification(nn.Module, SupportsCrossEncoding,
self._pooler = CrossEncodingPooler(config, self.classifier,
self.bert.pooler)
def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
self_weights = []