Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[tests] use torch_device instead of auto for model testing #29531

Merged
merged 5 commits into from
Mar 8, 2024
+3 −0
Merged

[tests] use torch_device instead of auto for model testing #29531

Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
167c188
use torch_device
faaany Mar 7, 2024
c4388f5
Merge branch 'huggingface:main' into auto
faaany Mar 8, 2024
c5e17af
skip for XPU
faaany Mar 8, 2024
77b763c
Merge branch 'auto' of https://github.com/faaany/transformers into auto
faaany Mar 8, 2024
5e7f979
Update tests/generation/test_utils.py
faaany Mar 8, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion tests/generation/test_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1073,6 +1073,9 @@ def test_beam_search_generate_dict_outputs_use_cache(self):
@require_torch_multi_accelerator
def test_model_parallel_beam_search(self):
for model_class in self.all_generative_model_classes:
if torch_device == "xpu":
faaany marked this conversation as resolved.
Show resolved Hide resolved
return unittest.skip("device_map='auto' does not work with XPU devices")

if model_class._no_split_modules is None:
continue

Expand All @@ -1081,7 +1084,7 @@ def test_model_parallel_beam_search(self):
model = model_class(config).eval()
with tempfile.TemporaryDirectory() as tmp_dir:
model.cpu().save_pretrained(tmp_dir)
new_model = model_class.from_pretrained(tmp_dir, device_map=torch_device)
new_model = model_class.from_pretrained(tmp_dir, device_map="auto")

new_model.generate(
input_ids,
Expand Down