Skip to content

[DONT MERGE] PR to debug CI failures on windows #6195

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

Closed
wants to merge 11 commits into from
Closed

[DONT MERGE] PR to debug CI failures on windows #6195

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
a0569b1
Skip big model in both cpu and cuda
YosuaMichael Jun 23, 2022
508951c
Test using nightly 20220621
YosuaMichael Jun 23, 2022
1eae59a
Try modify so windows also use older nightly
YosuaMichael Jun 23, 2022
a84e47f
Update config.yml
vfdev-5 Jun 23, 2022
d4c6fba
Update config.yml.in
vfdev-5 Jun 23, 2022
e76071e
Update install.sh
vfdev-5 Jun 23, 2022
de3769c
Skip only backprop
YosuaMichael Jun 23, 2022
7eb021f
Merge branch 'debug/check-ci-failure-on-big-model' of github.com:Yosu…
YosuaMichael Jun 23, 2022
5ec37f8
Change back the windows pytorch version to 20220621 to check if skipp…
YosuaMichael Jun 23, 2022
ce85914
Skip jit, fx, and backprop test for big model
YosuaMichael Jun 23, 2022
a0bae4d
Test the core 18 June again
YosuaMichael Jun 23, 2022
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 .circleci/config.yml
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion .circleci/config.yml.in
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -146,7 +146,7 @@ commands:
default: true
steps:
- pip_install:
args: --pre torch --extra-index-url https://download.pytorch.org/whl/nightly/cpu
args: --pre torch==1.13.0.dev20220618 --extra-index-url https://download.pytorch.org/whl/nightly/cpu
descr: Install PyTorch from nightly releases
- pip_install:
args: --no-build-isolation <<# parameters.editable >> --editable <</ parameters.editable >> .
Expand Down
3 changes: 2 additions & 1 deletion .circleci/unittest/windows/scripts/install.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,8 @@ else
fi

printf "Installing PyTorch with %s\n" "${cudatoolkit}"
conda install -y -c "pytorch-${UPLOAD_CHANNEL}" -c nvidia "pytorch-${UPLOAD_CHANNEL}"::pytorch[build="*${version}*"] "${cudatoolkit}"
# conda install -y -c "pytorch-${UPLOAD_CHANNEL}" -c nvidia "pytorch-${UPLOAD_CHANNEL}"::pytorch[build="*${version}*"] "${cudatoolkit}"
pip install --pre torch==1.13.0.dev20220618+cpu --extra-index-url https://download.pytorch.org/whl/nightly/cpu

torch_cuda=$(python -c "import torch; print(torch.cuda.is_available())")
echo torch.cuda.is_available is $torch_cuda
Expand Down
12 changes: 9 additions & 3 deletions test/test_models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -603,8 +603,8 @@ def test_classification_model(model_fn, dev):
"input_shape": (1, 3, 224, 224),
}
model_name = model_fn.__name__
if dev == "cuda" and SKIP_BIG_MODEL and model_name in skipped_big_models:
pytest.skip("Skipped to reduce memory usage. Set env var SKIP_BIG_MODEL=0 to enable test for this model")
# if dev == "cuda" and SKIP_BIG_MODEL and model_name in skipped_big_models:
# pytest.skip("Skipped to reduce memory usage. Set env var SKIP_BIG_MODEL=0 to enable test for this model")
kwargs = {**defaults, **_model_params.get(model_name, {})}
num_classes = kwargs.get("num_classes")
input_shape = kwargs.pop("input_shape")
Expand All @@ -613,9 +613,15 @@ def test_classification_model(model_fn, dev):
model.eval().to(device=dev)
# RNG always on CPU, to ensure x in cuda tests is bitwise identical to x in cpu tests
x = torch.rand(input_shape).to(device=dev)
out = model(x)
with torch.inference_mode():
out = model(x)
_assert_expected(out.cpu(), model_name, prec=0.1)
assert out.shape[-1] == num_classes

if SKIP_BIG_MODEL and model_name in skipped_big_models:
# Skip backprop test only
return

_check_jit_scriptable(model, (x,), unwrapper=script_model_unwrapper.get(model_name, None), eager_out=out)
_check_fx_compatible(model, x, eager_out=out)

Expand Down