SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Rename supports_static_cache to can_compile_fullgraph (#39505)

Browse Source 
* update all

* Apply suggestions from code review

Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>

* apply suggestions

* fix copies

---------

Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>
This commit is contained in:
Raushan Turganbay
2025-07-23 11:35:18 +02:00
committed by GitHub
parent b357cbb19d
commit eb1a007f7f
130 changed files with 143 additions and 148 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
src/transformers/modeling_utils.py
View File

@@ -2063,8 +2063,7 @@ class PreTrainedModel(nn.Module, EmbeddingAccessMixin, ModuleUtilsMixin, PushToH
# Flex Attention support
_supports_flex_attn = False
# Has support `torch.compile(fullgraph=True)`
_supports_static_cache = False
_can_compile_fullgraph = False
# A tensor parallel plan to be applied to the model when TP is enabled. For
# top-level models, this attribute is currently defined in respective model