diff options
| author | Jan Eilers <jan.eilers@arm.com> | 2020-11-20 11:59:40 +0000 |
|---|---|---|
| committer | Francis Murtagh <francis.murtagh@arm.com> | 2020-11-20 16:45:23 +0000 |
| commit | 225fb41627cdfa7f81406e22527531af1aeddcda (patch) | |
| tree | d0048a392d76bbaa104335848e4444125bbf83ff /delegate/src/test/ConvolutionTestHelper.hpp | |
| parent | 8bae82af919c4cddd638345a1d18fa6a689b4dbd (diff) | |
| download | armnn-225fb41627cdfa7f81406e22527531af1aeddcda.tar.gz | |
IVGCVSW-5550 Fix failing delegate Conv2d tests for CpuAcc/GpuAcc
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: Ibfd1968a622c2538bbfacf4802cd45096de59db5
Diffstat (limited to 'delegate/src/test/ConvolutionTestHelper.hpp')
| -rw-r--r-- | delegate/src/test/ConvolutionTestHelper.hpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/delegate/src/test/ConvolutionTestHelper.hpp b/delegate/src/test/ConvolutionTestHelper.hpp index b7705cc904..b3175174d2 100644 --- a/delegate/src/test/ConvolutionTestHelper.hpp +++ b/delegate/src/test/ConvolutionTestHelper.hpp @@ -91,7 +91,7 @@ std::vector<char> CreateConv2dTfLiteModel(tflite::BuiltinOperator convolutionOpe filterQuantizationParameters); auto biasTensorType = ::tflite::TensorType_FLOAT32; - if (tensorType == ::tflite::TensorType_UINT8) + if (tensorType == ::tflite::TensorType_INT8 || tensorType == ::tflite::TensorType_UINT8) { biasTensorType = ::tflite::TensorType_INT32; } |