commit | 8910528b126fed87628c23f7e51f7e018b13989a | [log] [tgz] |
---|---|---|
author | Matthew Bentham <matthew.bentham@arm.com> | Tue Nov 20 14:33:33 2018 +0000 |
committer | Matteo Martincigh <matteo.martincigh@arm.com> | Tue Nov 20 15:38:24 2018 +0000 |
tree | 67d7a3b0cfe8d0878f545d45b48c45c1e62320b2 | |
parent | 07fefa5d487bbd47c7755a7bb88b05edbe1d8342 [diff] [blame] |
IVGCVSW-1199 Disable auto-flattening of Compute Library tensors This is one of the reasons why the tests in https://review.mlplatform.org/#/c/ml/armnn/+/237/ are failing (but not the only reason). Change-Id: If485bade2a6dd013cba826cec71d748fc7747249
diff --git a/src/backends/aclCommon/ArmComputeTensorUtils.cpp b/src/backends/aclCommon/ArmComputeTensorUtils.cpp index 9fb78f8..ba7cd14 100644 --- a/src/backends/aclCommon/ArmComputeTensorUtils.cpp +++ b/src/backends/aclCommon/ArmComputeTensorUtils.cpp
@@ -40,7 +40,7 @@ for (unsigned int i = 0; i < tensorShape.GetNumDimensions(); i++) { // Note that our dimensions are stored in the opposite order to ACL's. - shape.set(tensorShape.GetNumDimensions() - i - 1, tensorShape[i]); + shape.set(tensorShape.GetNumDimensions() - i - 1, tensorShape[i], false); // TensorShape::set() flattens leading ones, so that batch size 1 cannot happen. // arm_compute tensors expect this.