diff options
| author | Ryan OShea <ryan.oshea3@arm.com> | 2022-11-07 16:20:48 +0000 |
|---|---|---|
| committer | ryan.oshea3 <ryan.oshea3@arm.com> | 2022-11-16 15:22:50 +0000 |
| commit | 31441595009182c985dacbedc70c41ee6664d070 (patch) | |
| tree | 248a85295aeff4022c9b395fc97748b0a0aa6b35 /src/backends/backendsCommon/WorkloadData.cpp | |
| parent | bd18eab07a8f30492de1e462b1815189014cb8d5 (diff) | |
| download | armnn-31441595009182c985dacbedc70c41ee6664d070.tar.gz | |
IVGCVSW-7214 Disable BF16-Turbo-Mode and remove conversion layers
- Remove Bf16ToFp32 Conversion Layer
- Remove Fp32ToBf16 Conversion Layer
- Remove B16 Conversion tests
* Throw exception if m_ReduceFp32ToBf16 optimzer option is set to true
* Provide comments to enable fast math in order to use bf16
* Update docs to inform users to enable fast math for bf16
Execute Network Changes
* Require bf16_turbo_mode to also have fast_math_enabled set to true
- Remove setting m_ReduceFp32ToBf16 optimizer option
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: Ibaa6da9d29c96a1ce32ff5196b0847fde9f04a1c
Diffstat (limited to 'src/backends/backendsCommon/WorkloadData.cpp')
| -rw-r--r-- | src/backends/backendsCommon/WorkloadData.cpp | 46 |
1 files changed, 0 insertions, 46 deletions
diff --git a/src/backends/backendsCommon/WorkloadData.cpp b/src/backends/backendsCommon/WorkloadData.cpp index 753fe06edb..62dfc6a38b 100644 --- a/src/backends/backendsCommon/WorkloadData.cpp +++ b/src/backends/backendsCommon/WorkloadData.cpp @@ -2222,52 +2222,6 @@ void LstmQueueDescriptor::Validate(const WorkloadInfo& workloadInfo) const } } -void ConvertBf16ToFp32QueueDescriptor::Validate(const WorkloadInfo& workloadInfo) const -{ - const std::string descriptorName{"ConvertBf16ToFp32QueueDescriptor"}; - - ValidateNumInputs(workloadInfo, descriptorName, 1); - ValidateNumOutputs(workloadInfo, descriptorName, 1); - - const TensorInfo& inputTensorInfo = workloadInfo.m_InputTensorInfos[0]; - const TensorInfo& outputTensorInfo = workloadInfo.m_OutputTensorInfos[0]; - - if (inputTensorInfo.GetDataType() != DataType::BFloat16) - { - throw InvalidArgumentException(descriptorName + ": Input tensor type must be BFloat16."); - } - - if (outputTensorInfo.GetDataType() != DataType::Float32) - { - throw InvalidArgumentException(descriptorName + ": Output tensor type must be Float32."); - } - - ValidateTensorShapesMatch(inputTensorInfo, outputTensorInfo, descriptorName, "input", "output"); -} - -void ConvertFp32ToBf16QueueDescriptor::Validate(const WorkloadInfo& workloadInfo) const -{ - const std::string descriptorName{"ConvertFp32ToBf16QueueDescriptor"}; - - ValidateNumInputs(workloadInfo, descriptorName, 1); - ValidateNumOutputs(workloadInfo, descriptorName, 1); - - const TensorInfo& inputTensorInfo = workloadInfo.m_InputTensorInfos[0]; - const TensorInfo& outputTensorInfo = workloadInfo.m_OutputTensorInfos[0]; - - if (inputTensorInfo.GetDataType() != DataType::Float32) - { - throw InvalidArgumentException(descriptorName + ": Input tensor type must be Float32."); - } - - if (outputTensorInfo.GetDataType() != DataType::BFloat16) - { - throw InvalidArgumentException(descriptorName + ": Output tensor type must be BFloat16."); - } - - ValidateTensorShapesMatch(inputTensorInfo, outputTensorInfo, descriptorName, "input", "output"); -} - void ConvertFp32ToFp16QueueDescriptor::Validate(const WorkloadInfo& workloadInfo) const { const std::string descriptorName{"ConvertFp32ToFp16QueueDescriptor"}; |