From a1b1e41bb261f5613f443fed7071936a360686ed Mon Sep 17 00:00:00 2001 From: Mohammed Suhail Munshi Date: Thu, 23 Mar 2023 22:21:31 +0000 Subject: Implement MatMul Function and Operator with Floating Point support for CPU - Implements MatMul function and operator for floating point datatype FP16/FP32 - Includes support for transposing dynamic tensors prior to matrix multiplication. - Adds tests for 2D/3D/4D+ tensors in MatMul with F32/F16 datatype (with all combinations of transposed/not-transposed tensors) - Updates fixture to allow for testing fused activation in MatMul - Adds tests for matmul with and without fused activation Resolved: [COMPMID-5898] Signed-off-by: Mohammed Suhail Munshi Change-Id: Iefa84b26dd723c9a51e6c3f91023152c6c31ace2 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9411 Reviewed-by: SiCong Li Tested-by: Arm Jenkins Benchmark: Arm Jenkins --- src/runtime/CL/functions/CLMatMul.cpp | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'src/runtime/CL/functions') diff --git a/src/runtime/CL/functions/CLMatMul.cpp b/src/runtime/CL/functions/CLMatMul.cpp index f42e4ff309..ae5a01f679 100644 --- a/src/runtime/CL/functions/CLMatMul.cpp +++ b/src/runtime/CL/functions/CLMatMul.cpp @@ -42,14 +42,16 @@ CLMatMul::CLMatMul() CLMatMul::~CLMatMul() = default; -void CLMatMul::configure(ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info) +void CLMatMul::configure(ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings) { + ARM_COMPUTE_UNUSED(settings); configure(CLKernelLibrary::get().get_compile_context(), lhs, rhs, output, matmul_info); } -void CLMatMul::configure(const CLCompileContext &compile_context, ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info) +void CLMatMul::configure(const CLCompileContext &compile_context, ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings) { ARM_COMPUTE_ERROR_ON_NULLPTR(lhs, rhs, output); + ARM_COMPUTE_UNUSED(settings); _impl->op = std::make_unique(); _impl->op->configure(compile_context, lhs->info(), rhs->info(), output->info(), matmul_info); -- cgit v1.2.1