apeforest commented on a change in pull request #15098: Add epsilon for
tolerance level to fix flaky test random_resize_crop
URL: https://github.com/apache/incubator-mxnet/pull/15098#discussion_r289502198
##########
File path: tests/python/unittest/test_image.py
##########
@@ -356,17 +356,18 @@ def test_det_augmenters(self):
pass
@with_seed()
- @unittest.skip('Flaky test. Skipping until a solution can be found.
Tracked by https://github.com/apache/incubator-mxnet/issues/14718')
def test_random_size_crop(self):
# test aspect ratio within bounds
width = np.random.randint(100, 500)
height = np.random.randint(100, 500)
src = np.random.rand(height, width, 3) * 255.
ratio = (0.75, 1)
+ epsilon = 0.05
Review comment:
Thanks @stu1130 for the discussion offline. The randomness is caused by sqrt
of area / or * ratio. Same epsilon is used in pytorch test as well. Approve
this change thereby.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
