diff options
author | Peter Hawkins <phawkins@google.com> | 2017-06-16 10:32:29 -0700 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-06-16 10:36:33 -0700 |
commit | a66de1eca225bc95e7972974a7089d84df8a8055 (patch) | |
tree | 7a332e8fd1f1656a0ef0caacbfff0851ab13967f /tensorflow/compiler/tests/tensor_array_ops_test.py | |
parent | d06121593035687176d8b660b83bab568853deff (diff) |
[TF:XLA] Refactor handling of Resources (Variables and TensorArrays) in the XLA bridge.
* Rename "Variable" to "Resource" in many places where non-Variable resources might be used.
* Add kTensorArray to the XlaCompiler::Argument enum. Remove kUninitializedVariable and make "initialized" a separate boolean field.
* Add a kind field to XlaResource. Add checks that Variables are not used where TensorArrays are expected, and vice-versa.
* Clean ups to the TensorArray operators.
PiperOrigin-RevId: 159244478
Diffstat (limited to 'tensorflow/compiler/tests/tensor_array_ops_test.py')
-rw-r--r-- | tensorflow/compiler/tests/tensor_array_ops_test.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tensorflow/compiler/tests/tensor_array_ops_test.py b/tensorflow/compiler/tests/tensor_array_ops_test.py index 00a7358130..b3067be51d 100644 --- a/tensorflow/compiler/tests/tensor_array_ops_test.py +++ b/tensorflow/compiler/tests/tensor_array_ops_test.py @@ -335,7 +335,7 @@ class TensorArrayTest(xla_test.XLATestCase): r0_bad = gen_data_flow_ops._tensor_array_read_v3( handle=w0.handle, index=0, dtype=dtypes.float64, flow_in=w0.flow) with self.assertRaisesOpError( - "TensorArray dtype is float but Op requested dtype double."): + "TensorArray dtype is float but op has dtype double."): r0_bad.eval() # Test reading from a different index than the one we wrote to |