marcoabreu closed pull request #12429: Adjust tolerance levels of
test_l2_normalization
URL: https://github.com/apache/incubator-mxnet/pull/12429
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):
diff --git a/tests/python/unittest/test_operator.py
b/tests/python/unittest/test_operator.py
index f246689e34e..2f94eb0aae1 100644
--- a/tests/python/unittest/test_operator.py
+++ b/tests/python/unittest/test_operator.py
@@ -3115,11 +3115,9 @@ def check_l2_normalization(in_shape, mode, dtype,
norm_eps=1e-10):
# compare numpy + mxnet
assert_almost_equal(exe.outputs[0].asnumpy(), np_out, rtol=1e-2 if dtype
is 'float16' else 1e-5, atol=1e-5)
# check gradient
- check_numeric_gradient(out, [in_data], numeric_eps=1e-3, rtol=1e-2,
atol=1e-3)
+ check_numeric_gradient(out, [in_data], numeric_eps=1e-3, rtol=1e-2,
atol=5e-3)
-# @haojin2: getting rid of the fixed seed as the flakiness could not be
reproduced.
-# tracked at: https://github.com/apache/incubator-mxnet/issues/11717
@with_seed()
def test_l2_normalization():
for dtype in ['float16', 'float32', 'float64']:
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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
