COMPMID-2968: Add support for QASYMM8_SIGNED in CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel

Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Change-Id: I37e6e76dbd5546c0eaedfacd01ea905c37148e8a
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2861
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 91 insertions, 0 deletions

diff --git a/arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.h b/arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.h
new file mode 100644
index 0000000000..439f569d07
--- /dev/null
+++ b/arm_compute/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.h
@@ -0,0 +1,91 @@
+/*
+ * Copyright (c) 2020 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef ARM_COMPUTE_CLGEMMLOWPQUANTIZEDOWNINT32SCALEBYFLOATKERNEL_H
+#define ARM_COMPUTE_CLGEMMLOWPQUANTIZEDOWNINT32SCALEBYFLOATKERNEL_H
+
+#include "arm_compute/core/CL/ICLKernel.h"
+
+namespace arm_compute
+{
+// Forward declarations
+class ICLTensor;
+
+/** OpenCL kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8/QASYMM8_SIGNED
+ *
+ * This kernel takes a final int32 accumulator value (the output of @ref CLGEMMLowpMatrixMultiplyKernel), and processes it to obtain the final QASYMM8/QASYMM8_SIGNED value.
+ * The following computations will be performed by the kernel:
+ *
+ *  -# Compute fixed point multiplication between each entry of input by result_fixedpoint_multiplier
+ *  -# Add bias to final result if bias tensor is not a nullptr
+ *  -# Requantize
+ *  -# Add offset to each result
+ *  -# Clamp the value between the specified min and max bounds
+ *  -# Clamp the resulting int32 values to
+ *      - to the [0..255] range and cast to QASYMM8.
+ *      - to the [-128..127] range and cast to QASYMM8_SIGNED.
+ */
+class CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel : public ICLKernel
+{
+public:
+    /** Constructor */
+    CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel();
+    /** Prevent instances of this class from being copied (As this class contains pointers)*/
+    CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel(const CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel &) = delete;
+    /** Prevent instances of this class from being copied (As this class contains pointers)*/
+    CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel &operator=(const CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel &) = delete;
+    /** Allow instances of this class to be moved */
+    CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel(CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel &&) = default;
+    /** Allow instances of this class to be moved */
+    CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel &operator=(CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel &&) = default;
+    /** Initialise the kernel's input and output.
+     *
+     * @param[in]  input  Input tensor. Data type supported: S32
+     * @param[in]  bias   Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required.
+     *                    Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input.
+     * @param[out] output Output tensor. Data type supported: Data type supported: QASYMM8/QASYMM8_SIGNED
+     * @param[in]  info   Output stage info. Used to pass the quantized output data type
+     */
+    void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, const GEMMLowpOutputStageInfo *info);
+    /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel
+     *
+     * @param[in] input  Input tensor. Data type supported: S32
+     * @param[in] bias   Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required.
+     *                   Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input.
+     * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8/QASYMM8_SIGNED
+     * @param[in] info   Output stage info. Used to pass the quantized output data type
+     *
+     * @return a status
+     */
+    static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo *info);
+
+    // Inherited methods overridden:
+    void run(const Window &window, cl::CommandQueue &queue) override;
+
+private:
+    const ICLTensor *_input;
+    const ICLTensor *_bias;
+    ICLTensor       *_output;
+};
+} // namespace arm_compute
+#endif /* ARM_COMPUTE_CLGEMMLOWPQUANTIZEDOWNINT32SCALEBYFLOATKERNEL_H */