OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

[test] Skip testing of source_fn_stack in light of export changes (#165176)

Browse Source 
This is in regards to https://github.com/pytorch/pytorch/pull/164691
where we are inlining into nn modules, and therefore it is causing this
test to fail. The test here looks for node.name which is quite different
with inlining.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165176
Approved by: https://github.com/andrewor14
ghstack dependencies: #165172
This commit is contained in:
Animesh Jain 2025-10-10 12:54:09 -07:00 committed by PyTorch MergeBot
parent ef50c9b557
commit d73416642f
1 changed files with 0 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
test/quantization/pt2e/test_quantize_pt2e_qat.py
View File

@ -665,12 +665,6 @@ class TestQuantizePT2EQAT_ConvBn_Base(PT2EQATTestCase):
self.assertNotEqual(get_source_fn(second_conv), get_source_fn(second_relu))
self.assertNotEqual(get_source_fn(first_relu), get_source_fn(second_relu))
# Assert that "backbone" exists only in the second set of conv and relu's partition
self.assertTrue("backbone" not in get_source_fn(first_conv))
self.assertTrue("backbone" not in get_source_fn(first_relu))
self.assertTrue("backbone" in get_source_fn(second_conv))
self.assertTrue("backbone" in get_source_fn(second_relu))
def test_qat_conv_bn_bias_derived_qspec(self):
m = self._get_conv_bn_model()
example_inputs = self.example_inputs