xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-15 01:45:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Docs] Fix warnings in vllm/profiler and vllm/transformers_utils (#25220)

Browse Source 
Signed-off-by: windsonsea <haifeng.yao@daocloud.io>
This commit is contained in:
Michael Yao 2025-09-21 07:39:47 +08:00 committed by GitHub
parent bef180f009
commit 367a480bd3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
mkdocs.yaml
View File

@ -102,6 +102,7 @@ plugins:
- https://numpy.org/doc/stable/objects.inv - https://numpy.org/doc/stable/objects.inv
- https://pytorch.org/docs/stable/objects.inv - https://pytorch.org/docs/stable/objects.inv
- https://psutil.readthedocs.io/en/stable/objects.inv - https://psutil.readthedocs.io/en/stable/objects.inv
- https://huggingface.co/docs/transformers/main/en/objects.inv
markdown_extensions: markdown_extensions:
- attr_list - attr_list

4
vllm/profiler/layerwise_profile.py
View File

@ -353,8 +353,8 @@ class layerwise_profile(profile):
Args: Args:
num_running_seqs (Optional[int], optional): When given, num_running_seqs (Optional[int], optional): When given,
num_running_seqs will be passed to LayerProfileResults for metadata num_running_seqs will be passed to LayerProfileResults
update. Defaults to None. for metadata update. Defaults to None.
""" """
super().__init__( super().__init__(
activities=[ProfilerActivity.CPU, ProfilerActivity.CUDA], activities=[ProfilerActivity.CPU, ProfilerActivity.CUDA],

3
vllm/transformers_utils/configs/jais.py
View File

@ -74,8 +74,7 @@ class JAISConfig(PretrainedConfig):
use_cache (`bool`, *optional*, defaults to `True`): use_cache (`bool`, *optional*, defaults to `True`):
Whether or not the model should return the last key/values Whether or not the model should return the last key/values
attentions (not used by all models). attentions (not used by all models).
scale_attn_by_inverse_layer_idx scale_attn_by_inverse_layer_idx (`bool`, *optional*, default `True`):
(`bool`, *optional*, defaults to `False`):
Whether to additionally scale attention weights Whether to additionally scale attention weights
by `1 / layer_idx + 1`. by `1 / layer_idx + 1`.
reorder_and_upcast_attn (`bool`, *optional*, defaults to `False`): reorder_and_upcast_attn (`bool`, *optional*, defaults to `False`):