Skip to content

Fixed the auto device detection #1674

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

Merged
merged 3 commits into from
Mar 19, 2024
Merged

Fixed the auto device detection #1674

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
edc0879
update the device name to current device
yiliu30 Mar 15, 2024
319b071
udapte the UTs
yiliu30 Mar 15, 2024
7c8a612
fixed the ut
yiliu30 Mar 15, 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
2 changes: 1 addition & 1 deletion neural_compressor/torch/utils/environ.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -66,5 +66,5 @@ def get_device(device_name="auto"):
from neural_compressor.torch.utils.auto_accelerator import auto_detect_accelerator

runtime_accelerator = auto_detect_accelerator(device_name)
device = runtime_accelerator.name()
device = runtime_accelerator.current_device_name()
return device
11 changes: 11 additions & 0 deletions ...tion/weight_only/test_auto_accelerator.py → test/3x/torch/test_auto_accelerator.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
import pytest
import torch

from neural_compressor.torch.utils import get_device
from neural_compressor.torch.utils.auto_accelerator import accelerator_registry, auto_detect_accelerator


Expand Down Expand Up @@ -52,6 +53,16 @@ def test_cuda_accelerator(self, force_use_cuda):
assert accelerator.synchronize() is None
assert accelerator.empty_cache() is None

@pytest.mark.skipif(torch.cuda.device_count() < 2, reason="Only one GPU is available")
def test_get_device(self):
accelerator = auto_detect_accelerator()
assert accelerator.set_device(1) is None
assert accelerator.current_device_name() == "cuda:1"
cur_device = get_device()
assert cur_device == "cuda:1"
tmp_tensor = torch.tensor([1, 2], device=cur_device)
assert "cuda:1" == str(tmp_tensor.device)


class TestAutoAccelerator:

Expand Down