diff options
| author | arovir01 <Aron.Virginas-Tar@arm.com> | 2018-10-01 17:08:59 +0100 |
|---|---|---|
| committer | Matthew Bentham <matthew.bentham@arm.com> | 2018-10-10 16:16:57 +0100 |
| commit | 616e775763280992de92287b129dc335be91a24c (patch) | |
| tree | 5b7ced2fa8005f4371fe8005ee98cb88d16e0e09 /src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp | |
| parent | 29cadb3ff3066d6feb1801a921be326bf5ed8f47 (diff) | |
| download | armnn-616e775763280992de92287b129dc335be91a24c.tar.gz | |
IVGCVSW-1913: Fix for ValidationTest.concat_float_3_relaxed
* Added RefPermuteFloat16Workload to serve as a fallback when CL
does not support the required permute configuration for FP16
* Move Half.hpp to armnnUtils as the utils library should not be
including private header files from the armnn library
Change-Id: Ibf0f698451e8406f7ed7cce470dab60b6d16361d
Diffstat (limited to 'src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp')
| -rw-r--r-- | src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp b/src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp index 261de3d0ce..6bcf6e082c 100644 --- a/src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp +++ b/src/backends/neon/workloads/NeonConvertFp32ToFp16Workload.cpp @@ -5,9 +5,8 @@ #include "NeonConvertFp32ToFp16Workload.hpp" -#include <Half.hpp> +#include <armnnUtils/Half.hpp> #include <FloatingPointConverter.hpp> - #include <Profiling.hpp> #include <backends/WorkloadUtils.hpp> |