riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-08-01 18:51:14 +06:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

[tests] make test_sdpa_can_compile_dynamic device-agnostic (#32519)

Browse Source 
* enable

* fix
This commit is contained in:
Fanli Lin 2024-08-19 19:46:59 +08:00 committed by GitHub
parent 54b7703682
commit e55b33ceb4
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
tests/test_modeling_common.py
View File

@ -79,6 +79,7 @@ from transformers.testing_utils import (
require_read_token, require_read_token,
require_safetensors, require_safetensors,
require_torch, require_torch,
require_torch_accelerator,
require_torch_gpu, require_torch_gpu,
require_torch_multi_accelerator, require_torch_multi_accelerator,
require_torch_multi_gpu, require_torch_multi_gpu,
@ -4105,17 +4106,17 @@ class ModelTesterMixin:
_ = model(**inputs_dict) _ = model(**inputs_dict)
@require_torch_sdpa @require_torch_sdpa
@require_torch_gpu @require_torch_accelerator
@slow @slow
def test_sdpa_can_compile_dynamic(self): def test_sdpa_can_compile_dynamic(self):
if not self.has_attentions: if not self.has_attentions:
self.skipTest(reason="Model architecture does not support attentions") self.skipTest(reason="Model architecture does not support attentions")
if "cuda" in torch_device:
compute_capability = torch.cuda.get_device_capability()
major, _ = compute_capability
compute_capability = torch.cuda.get_device_capability() if not torch.version.cuda or major < 8:
major, _ = compute_capability self.skipTest(reason="This test requires an NVIDIA GPU with compute capability >= 8.0")
if not torch.version.cuda or major < 8:
self.skipTest(reason="This test requires an NVIDIA GPU with compute capability >= 8.0")
for model_class in self.all_model_classes: for model_class in self.all_model_classes:
if not model_class._supports_sdpa: if not model_class._supports_sdpa: