experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

linear_operator_toeplitz_test: Reduce flakes by increasing tolerance. Test

Browse Source 
has recently been flaky.
PiperOrigin-RevId: 294571109
Change-Id: Id9be7b954d97a7a040909f080859793d48e19bba
This commit is contained in:
Ian Langmore 2020-02-11 18:53:06 -08:00 committed by TensorFlower Gardener
parent 08eb6f9bea
commit 21a7c41fd1
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
tensorflow/python/kernel_tests/linalg/linear_operator_toeplitz_test.py
View File

@ -53,14 +53,14 @@ class LinearOperatorToeplitzTest(
def setUp(self):
# TODO(srvasude): Lower these tolerances once specialized solve and
# determinants are implemented.
self._atol[dtypes.float32] = 1e-5
self._rtol[dtypes.float32] = 1e-5
self._atol[dtypes.float64] = 1e-10
self._rtol[dtypes.float64] = 1e-10
self._atol[dtypes.complex64] = 1e-5
self._rtol[dtypes.complex64] = 1e-5
self._atol[dtypes.complex128] = 1e-10
self._rtol[dtypes.complex128] = 1e-10
self._atol[dtypes.float32] = 1e-4
self._rtol[dtypes.float32] = 1e-4
self._atol[dtypes.float64] = 1e-9
self._rtol[dtypes.float64] = 1e-9
self._atol[dtypes.complex64] = 1e-4
self._rtol[dtypes.complex64] = 1e-4
self._atol[dtypes.complex128] = 1e-9
self._rtol[dtypes.complex128] = 1e-9
@staticmethod
def skip_these_tests():