diff options
author | Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> | 2023-02-09 11:52:06 +0000 |
---|---|---|
committer | Mohmun02 <MohammedSuhail.Munshi@arm.com> | 2023-02-15 16:13:08 +0000 |
commit | ecaa10a5594b58f14f5962cdda71d9313d4f3aa8 (patch) | |
tree | 55a0ec0557710438bdece14739f0c145c6789725 /src | |
parent | 2cc6e0bcdcca5035670f5fec95c3621f619a8acd (diff) | |
download | ComputeLibrary-ecaa10a5594b58f14f5962cdda71d9313d4f3aa8.tar.gz |
Fix Intermittent Neon™ ReduceMean QASYMM8 Mismatch
- Dividing scale by number of elements causes accuracy loss due to limitations in float datatype and truncation to int
- Adds rounding after division on aarch64 to negate this.
Resolves: [COMPMID-5839]
Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com>
Change-Id: I54ef0f7e56f39da1fa5f30378f551b5ca419a61d
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/492456
Tested-by: bsgcomp <bsgcomp@arm.com>
Comments-Addressed: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9110
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src')
-rw-r--r-- | src/core/NEON/kernels/NEReductionOperationKernel.cpp | 28 |
1 files changed, 20 insertions, 8 deletions
diff --git a/src/core/NEON/kernels/NEReductionOperationKernel.cpp b/src/core/NEON/kernels/NEReductionOperationKernel.cpp index e0f43ab176..19955af493 100644 --- a/src/core/NEON/kernels/NEReductionOperationKernel.cpp +++ b/src/core/NEON/kernels/NEReductionOperationKernel.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2022 Arm Limited. + * Copyright (c) 2017-2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -625,13 +625,13 @@ struct RedOpX_quantized Iterator input(in, in_win_no_pad); Iterator output(out, out_window); - const float in_offset = static_cast<float>(iq_info.offset); + const auto in_offset = static_cast<float>(iq_info.offset); const float in_scale = iq_info.scale; - const float out_offset = static_cast<float>(oq_info.offset); + const auto out_offset = static_cast<float>(oq_info.offset); const float out_scale = oq_info.scale; - const float num_elements = static_cast<float>(in_info.dimension(0)); + const auto num_elements = static_cast<float>(in_info.dimension(0)); const float A = in_scale / (out_scale * num_elements); const float B = out_offset - (in_scale * in_offset) / (out_scale); @@ -1382,10 +1382,17 @@ struct RedOpYZW_quantized vec_res_value3_f = wrapper::vmla(vec_B, wrapper::vcvt<float>(vec_res_value3), vec_A); vec_res_value4_f = wrapper::vmla(vec_B, wrapper::vcvt<float>(vec_res_value4), vec_A); - vec_res_value1 = wrapper::vcvt<T>(vec_res_value1_f); - vec_res_value2 = wrapper::vcvt<T>(vec_res_value2_f); - vec_res_value3 = wrapper::vcvt<T>(vec_res_value3_f); - vec_res_value4 = wrapper::vcvt<T>(vec_res_value4_f); +#ifdef __aarch64__ + vec_res_value1 = wrapper::vcvta<PromotedType>(vec_res_value1_f); + vec_res_value2 = wrapper::vcvta<PromotedType>(vec_res_value2_f); + vec_res_value3 = wrapper::vcvta<PromotedType>(vec_res_value3_f); + vec_res_value4 = wrapper::vcvta<PromotedType>(vec_res_value4_f); +#else // defined(__aarch64__) + vec_res_value1 = wrapper::vcvt<PromotedType>(vec_res_value1_f); + vec_res_value2 = wrapper::vcvt<PromotedType>(vec_res_value2_f); + vec_res_value3 = wrapper::vcvt<PromotedType>(vec_res_value3_f); + vec_res_value4 = wrapper::vcvt<PromotedType>(vec_res_value4_f); +#endif // __aarch64__ const auto temp16x8t_1 = wrapper::vcombine(wrapper::vqmovn(vec_res_value1), wrapper::vqmovn(vec_res_value2)); const auto temp16x8t_2 = wrapper::vcombine(wrapper::vqmovn(vec_res_value3), wrapper::vqmovn(vec_res_value4)); @@ -1521,7 +1528,12 @@ struct RedOpYZW_quantized { case ReductionOperation::MEAN_SUM: { + // Apply previously calculated coefficients (with rounding on aarch64) +#ifdef __aarch64__ + const int32_t res = arm_compute::support::cpp11::round(A * (static_cast<float>(res_value_q)) + B); +#else // defined(__aarch64__) const int32_t res = A * (static_cast<float>(res_value_q)) + B; +#endif // __aarch64__ *reinterpret_cast<T *>(output.ptr() + x) = utils::cast::saturate_cast<T>(res); break; } |