diff options
author | Jakub Sujak <jakub.sujak@arm.com> | 2023-04-17 12:08:48 +0100 |
---|---|---|
committer | Omar Al Khatib <omar.alkhatib@arm.com> | 2023-05-05 14:48:28 +0000 |
commit | e9b3ee2badebf91188c1cd0e59d6aaa30ed60985 (patch) | |
tree | 750c39df7c0113caf6a893bb6af6e9ef1ecc3756 /arm_compute/runtime/NEON | |
parent | edafe7f5fdc056fddc395c70420fc869dcb7d9fb (diff) | |
download | ComputeLibrary-e9b3ee2badebf91188c1cd0e59d6aaa30ed60985.tar.gz |
Connect CLMatMul function to quantized kernels and resolve NE BatchMatMul int_8 failures
* Adapt the CLMatMul function and ClMatMul operator to use quantized kernels.
* Add function-level tests.
Resolves: COMPMID-5929 and COMPMID-5811
Change-Id: I5348cdcf07b8074c138e04dfef0a73399377accd
Signed-off-by: Jakub Sujak <jakub.sujak@arm.com>
Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9575
Reviewed-by: Mohmun02 <MohammedSuhail.Munshi@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/NEON')
-rw-r--r-- | arm_compute/runtime/NEON/functions/NEMatMul.h | 22 |
1 files changed, 12 insertions, 10 deletions
diff --git a/arm_compute/runtime/NEON/functions/NEMatMul.h b/arm_compute/runtime/NEON/functions/NEMatMul.h index 0f3e3adacc..a331c55a98 100644 --- a/arm_compute/runtime/NEON/functions/NEMatMul.h +++ b/arm_compute/runtime/NEON/functions/NEMatMul.h @@ -21,8 +21,8 @@ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ -#ifndef ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL -#define ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL +#ifndef ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL +#define ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL #include "arm_compute/runtime/IFunction.h" #include <memory> @@ -80,25 +80,27 @@ public: * - Any * * Valid data type configurations: - * |src0 |src1 |dst | + * |lhs |rhs |dst | * |:--------------|:------------------|:--------------| * |F32 |F32 |F32 | * |F16 |F16 |F16 | + * |QASYMM8_SIGNED |QASYMM8_SIGNED |QASYMM8_SIGNED | + * |QASYMM8 |QASYMM8 |QASYMM8 | * - * @param[in] lhs Input source tensor. - * @param[in] rhs Input source tensor. - * @param[out] output Output tensor. Data type supported: same as @p lhs/rhs - * @param[in] info Class containing flags to transpose lhs/rhs + * @param[in] lhs Left-hand side tensor info. Data types supported: F16/F32/QASYMM8_SIGNED/QASYMM8. + * @param[in] rhs Right-hand side tensor info. Data types supported: same as @p lhs. + * @param[out] dst Output tensor to store the result of the batched matrix multiplication. Data types supported: same as @p lhs / @p rhs. + * @param[in] info Contains MatMul operation information described in @ref MatMulInfo. * @param[in] settings Class containing flags for function level settings i.e fast math */ - void configure(ITensor *lhs, ITensor *rhs, ITensor *output, const MatMulInfo &info, const CpuMatMulSettings &settings); + void configure(ITensor *lhs, ITensor *rhs, ITensor *dst, const MatMulInfo &info, const CpuMatMulSettings &settings); /** Static function to check if given info will lead to a valid configuration of @ref NEMatMul * * Parameters are similar to @ref NEMatMul::configure() * * @return Status */ - static Status validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &info, const CpuMatMulSettings &settings); + static Status validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *dst, const MatMulInfo &info, const CpuMatMulSettings &settings); // Inherited methods overridden void run() override; @@ -108,4 +110,4 @@ private: std::unique_ptr<Impl> _impl; }; } -#endif /* ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL */ +#endif /* ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL */ |