Skip to content

Skip int4 QAT tests for nightly for now #521

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 1 commit into from
Jul 17, 2024
Merged

Skip int4 QAT tests for nightly for now #521

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
11 changes: 10 additions & 1 deletion test/quantization/test_qat.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,10 @@
get_groupwise_affine_qparams,
groupwise_affine_quantize_tensor,
)
from torchao.utils import TORCH_VERSION_AFTER_2_4
from torchao.utils import (
TORCH_VERSION_AFTER_2_4,
TORCH_VERSION_AFTER_2_5,
)


# TODO: put this in a common test utils file
Expand Down Expand Up @@ -366,6 +369,8 @@ def _assert_close_4w(self, val, ref):

@unittest.skipIf(not TORCH_VERSION_AFTER_2_4, "skipping when torch version is 2.4 or lower")
@unittest.skipIf(not _CUDA_IS_AVAILABLE, "skipping when cuda is not available")
# TODO: remove once we fix int4 error: https://github.com/pytorch/ao/pull/517
@unittest.skipIf(TORCH_VERSION_AFTER_2_5, "int4 doesn't work for 2.5+ right now")
def test_qat_4w_primitives(self):
n_bit = 4
group_size = 32
Expand Down Expand Up @@ -411,6 +416,8 @@ def test_qat_4w_primitives(self):

@unittest.skipIf(not TORCH_VERSION_AFTER_2_4, "skipping when torch version is 2.4 or lower")
@unittest.skipIf(not _CUDA_IS_AVAILABLE, "skipping when cuda is not available")
# TODO: remove once we fix int4 error: https://github.com/pytorch/ao/pull/517
@unittest.skipIf(TORCH_VERSION_AFTER_2_5, "int4 doesn't work for 2.5+ right now")
def test_qat_4w_linear(self):
from torchao.quantization.prototype.qat import Int4WeightOnlyQATLinear
from torchao.quantization.GPTQ import WeightOnlyInt4Linear
Expand Down Expand Up @@ -439,6 +446,8 @@ def test_qat_4w_linear(self):

@unittest.skipIf(not TORCH_VERSION_AFTER_2_4, "skipping when torch version is 2.4 or lower")
@unittest.skipIf(not _CUDA_IS_AVAILABLE, "skipping when cuda is not available")
# TODO: remove once we fix int4 error: https://github.com/pytorch/ao/pull/517
@unittest.skipIf(TORCH_VERSION_AFTER_2_5, "int4 doesn't work for 2.5+ right now")
def test_qat_4w_quantizer(self):
from torchao.quantization.prototype.qat import Int4WeightOnlyQATQuantizer
from torchao.quantization.GPTQ import Int4WeightOnlyQuantizer
Expand Down
Loading