diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2017-12-11 13:48:51 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-12-11 13:53:21 -0800 |
| commit | dd77f385591c8b6ef7ab8dae7429c7eff7813a1e (patch) | |
| tree | 462c80b1f4bc994aecfc1b595b255a6f557a506c /tensorflow/compiler/xla/service/shape_inference_test.cc | |
| parent | db198b8618776731cc2871369f057b176488a1fe (diff) | |
[XLA] Move BatchDot unrolling from TF2XLA bridge to AlgebraicSimplifier so that unrolling can be selectively enabled/disabled per backend (should be no performance change).
PiperOrigin-RevId: 178666990
Diffstat (limited to 'tensorflow/compiler/xla/service/shape_inference_test.cc')
| -rw-r--r-- | tensorflow/compiler/xla/service/shape_inference_test.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tensorflow/compiler/xla/service/shape_inference_test.cc b/tensorflow/compiler/xla/service/shape_inference_test.cc index 7af2805f12..99d87f3b55 100644 --- a/tensorflow/compiler/xla/service/shape_inference_test.cc +++ b/tensorflow/compiler/xla/service/shape_inference_test.cc @@ -1062,7 +1062,7 @@ TEST_F(ShapeInferenceTest, DotWithMisatchedBatchDimNumbersFails) { ShapeInference::InferDotOpShape(lhs_shape, rhs_shape, dot_dnums); ASSERT_FALSE(inferred_status.ok()); ASSERT_THAT(inferred_status.status().error_message(), - HasSubstr("batch dimension numbers and sizes must match")); + HasSubstr("batch dimension numbers must precede non-batch")); } // BatchMatMul with out-of-range dimension numbers fails. |