frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 12:54:11 +08:00
Code Issues Packages Projects Releases Wiki Activity

[cuDNN][TF32] Account for TF32 in test_super_resolution_cuda (#161662)

Browse Source 
cuDNN seems to be dispatching to TF32 kernels on B200

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161662
Approved by: https://github.com/Skylion007
This commit is contained in:
Eddie Yan
2025-08-28 08:42:30 +00:00
committed by PyTorch MergeBot
parent 196232bb93
commit 2e77a08b95
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/jit/test_models.py
View File

@ -7,6 +7,7 @@ import unittest
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.testing._internal.common_cuda import tf32_on_and_off
from torch.testing._internal.common_utils import (
enable_profiling_mode_for_profiling_tests,
GRAPH_EXECUTOR,
@ -482,6 +483,7 @@ class TestModels(JitTestCase):
self._test_super_resolution(self, device="cpu")
@unittest.skipIf(not RUN_CUDA, "no CUDA")
@tf32_on_and_off(0.02)
def test_super_resolution_cuda(self):
# XXX: export_import on CUDA modules doesn't work (#11480)
self._test_super_resolution(self, device="cuda", check_export_import=False)