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
diff --git a/tests/ExecuteNetwork/ExecuteNetworkParams.cpp b/tests/ExecuteNetwork/ExecuteNetworkParams.cpp
index 155a4c4..fa467c9 100644
--- a/tests/ExecuteNetwork/ExecuteNetworkParams.cpp
+++ b/tests/ExecuteNetwork/ExecuteNetworkParams.cpp
@@ -60,10 +60,9 @@
}
CheckClTuningParameter(m_TuningLevel, m_TuningPath, m_ComputeDevices);
- if (m_EnableBf16TurboMode && m_EnableFp16TurboMode)
+ if (m_EnableBf16TurboMode && !m_EnableFastMath)
{
- throw armnn::InvalidArgumentException("BFloat16 and Float16 turbo mode cannot be "
- "enabled at the same time.");
+ throw armnn::InvalidArgumentException("To use BF16 please use --enable-fast-math. ");
}
// Check input tensor shapes
@@ -124,7 +123,6 @@
armnn::OptimizerOptions options;
options.m_ReduceFp32ToFp16 = m_EnableFp16TurboMode;
- options.m_ReduceFp32ToBf16 = m_EnableBf16TurboMode;
options.m_Debug = m_PrintIntermediate;
options.m_DebugToFile = m_PrintIntermediateOutputsToFile;
options.m_ProfilingEnabled = m_EnableProfiling;