Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 1 | /* |
Manuel Bottini | 9c9b70b | 2019-07-01 17:35:56 +0100 | [diff] [blame] | 2 | * Copyright (c) 2017-2019 ARM Limited. |
Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 3 | * |
| 4 | * SPDX-License-Identifier: MIT |
| 5 | * |
| 6 | * Permission is hereby granted, free of charge, to any person obtaining a copy |
| 7 | * of this software and associated documentation files (the "Software"), to |
| 8 | * deal in the Software without restriction, including without limitation the |
| 9 | * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or |
| 10 | * sell copies of the Software, and to permit persons to whom the Software is |
| 11 | * furnished to do so, subject to the following conditions: |
| 12 | * |
| 13 | * The above copyright notice and this permission notice shall be included in all |
| 14 | * copies or substantial portions of the Software. |
| 15 | * |
| 16 | * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR |
| 17 | * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, |
| 18 | * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE |
| 19 | * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER |
| 20 | * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, |
| 21 | * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE |
| 22 | * SOFTWARE. |
| 23 | */ |
Manuel Bottini | 1f332d4 | 2019-11-29 17:25:25 +0000 | [diff] [blame] | 24 | #ifndef ARM_COMPUTE_CLGEMMLOWPOUTPUTSTAGE_H |
| 25 | #define ARM_COMPUTE_CLGEMMLOWPOUTPUTSTAGE_H |
Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 26 | |
| 27 | #include "arm_compute/runtime/CL/ICLSimpleFunction.h" |
| 28 | |
| 29 | /** This file contains all available output stages for GEMMLowp on OpenCL. |
| 30 | * |
| 31 | * In gemmlowp, the "output stage" is the process that takes a final int32 accumulator value (the output of @ref CLGEMMLowpMatrixMultiplyCore), |
Manuel Bottini | 1f332d4 | 2019-11-29 17:25:25 +0000 | [diff] [blame] | 32 | * and processes it to obtain the final QASYMM8/QASYMM8_SIGNED value. |
Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 33 | * |
| 34 | * More information about the GEMMLowp output stage can be found at https://github.com/google/gemmlowp/blob/master/doc/output.md |
| 35 | */ |
| 36 | |
| 37 | namespace arm_compute |
| 38 | { |
| 39 | class ITensor; |
| 40 | |
| 41 | /** Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8Scale on OpenCL. |
| 42 | * |
| 43 | * CLGEMMLowpQuantizeDownInt32ToUint8Scale depends on 3 parameters: result_offset, result_mult_int, result_shift |
| 44 | * The final result is: |
| 45 | * |
| 46 | * ((input[i][k] + result_offset) * result_mult_int) >> result_shift |
| 47 | * |
| 48 | * In case the bias tensor is provided, the final result is: |
| 49 | * |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 50 | * ((input[i][k] + bias[k] + result_offset) * result_mult_int) >> result_shift |
Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 51 | * |
| 52 | * This function calls the following OpenCL kernels: |
| 53 | * |
| 54 | * -# @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleKernel |
| 55 | * |
| 56 | * @note The function accepts also 2 optional input arguments (min and max) which can be used to implement "rectified linear unit" activation functions |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 57 | * after the result is shifted right by result_shift |
Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 58 | */ |
| 59 | class CLGEMMLowpQuantizeDownInt32ToUint8Scale : public ICLSimpleFunction |
| 60 | { |
| 61 | public: |
| 62 | /** Initialise the kernel's inputs, output |
| 63 | * |
| 64 | * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
| 65 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. |
| 66 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 67 | * @param[out] output Output tensor. Data type supported: Data type supported: QASYMM8 |
| 68 | * @param[in] result_offset Offset to be added to each element of the input matrix |
| 69 | * @param[in] result_mult_int Value to be multiplied to each element of the input matrix when once the result_offset has been add |
| 70 | * @param[in] result_shift Number of bits to shift right the result before converting back to QASYMM8 |
| 71 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
| 72 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, |
| 73 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
| 74 | */ |
| 75 | void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_offset, int result_mult_int, int result_shift, int min = 0, int max = 0); |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 76 | /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpQuantizeDownInt32ToUint8Scale |
| 77 | * |
| 78 | * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
| 79 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. |
| 80 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 81 | * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8 |
| 82 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
| 83 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, |
| 84 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
| 85 | * |
Georgios Pinitas | 631c41a | 2017-12-06 11:53:03 +0000 | [diff] [blame] | 86 | * @return a status |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 87 | */ |
Georgios Pinitas | 631c41a | 2017-12-06 11:53:03 +0000 | [diff] [blame] | 88 | static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0); |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 89 | }; |
| 90 | |
| 91 | /** Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint on OpenCL. |
| 92 | * |
| 93 | * CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint depends on 3 parameters: |
| 94 | * |
| 95 | * result_fixedpoint_multiplier, result_shift, result_offset_after_shift |
| 96 | * |
| 97 | * The final result is: |
| 98 | * |
| 99 | * (FixedPointMul(input[i][k], result_fixedpoint_multiplier) >> result_shift) + result_offset_after_shift |
| 100 | * |
| 101 | * where FixedPointMul(x, y) is the nearest integer to the following |
| 102 | * mathematical expression, evaluated without overflow or intermediate rounding: |
| 103 | * |
| 104 | * (x * y) / 2^31 |
| 105 | * |
| 106 | * For more information: https://github.com/google/gemmlowp/blob/master/public/output_stages.h#L68 |
| 107 | * |
| 108 | * In case the bias tensor is provided, the final result is: |
| 109 | * |
| 110 | * ((FixedPointMul(input[i][k] + bias[k], result_fixedpoint_multiplier)) >> result_shift) + result_offset_after_shift |
| 111 | * |
| 112 | * This function calls the following OpenCL kernels: |
| 113 | * |
| 114 | * -# @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel |
| 115 | * |
| 116 | * @note The function accepts also 2 optional input arguments (min and max) which can be used to implement "rectified linear unit" activation functions |
| 117 | * after the result is shifted right by result_shift |
| 118 | */ |
| 119 | class CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint : public ICLSimpleFunction |
| 120 | { |
| 121 | public: |
| 122 | /** Initialise the kernel's inputs, output |
| 123 | * |
| 124 | * @param[in] input Input tensor. Data type supported: S32 |
| 125 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required. |
| 126 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 127 | * @param[out] output Output tensor. Data type supported: Data type supported: QASYMM8 |
| 128 | * @param[in] result_fixedpoint_multiplier Fixed point value to be multiplied to each element of the input matrix when once the result_offset has been add |
| 129 | * @param[in] result_shift Number of bits to shift right the result after the fixed point multiplication |
| 130 | * @param[in] result_offset_after_shift Offset to be applied to result before converting it back to QASYMM8 |
| 131 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
| 132 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, |
| 133 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
| 134 | */ |
Georgios Pinitas | 932491f | 2018-09-21 16:33:15 +0100 | [diff] [blame] | 135 | void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, |
Gian Marco Iodice | 4b90865 | 2018-10-18 10:21:02 +0100 | [diff] [blame] | 136 | int min = 0, int max = 0); |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 137 | /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint |
| 138 | * |
Gian Marco Iodice | 4b90865 | 2018-10-18 10:21:02 +0100 | [diff] [blame] | 139 | * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
| 140 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. |
| 141 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 142 | * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8 |
| 143 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
| 144 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, |
Georgios Pinitas | 932491f | 2018-09-21 16:33:15 +0100 | [diff] [blame] | 145 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 146 | * |
Georgios Pinitas | 631c41a | 2017-12-06 11:53:03 +0000 | [diff] [blame] | 147 | * @return a status |
Gian Marco | 58c5794 | 2017-11-28 09:10:03 +0000 | [diff] [blame] | 148 | */ |
Gian Marco Iodice | 4b90865 | 2018-10-18 10:21:02 +0100 | [diff] [blame] | 149 | static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0); |
Gian Marco | 05288a2 | 2017-11-21 10:57:50 +0000 | [diff] [blame] | 150 | }; |
Georgios Pinitas | 51e53a3 | 2018-10-22 13:49:08 +0100 | [diff] [blame] | 151 | |
Manuel Bottini | 1f332d4 | 2019-11-29 17:25:25 +0000 | [diff] [blame] | 152 | /** Basic function to execute CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPoint on OpenCL. |
| 153 | * |
| 154 | * CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPoint depends on 3 parameters: |
| 155 | * |
| 156 | * result_fixedpoint_multiplier, result_shift, result_offset_after_shift |
| 157 | * |
| 158 | * The final result is: |
| 159 | * |
| 160 | * (FixedPointMul(input[i][k], result_fixedpoint_multiplier) >> result_shift) + result_offset_after_shift |
| 161 | * |
| 162 | * where FixedPointMul(x, y) is the nearest integer to the following |
| 163 | * mathematical expression, evaluated without overflow or intermediate rounding: |
| 164 | * |
| 165 | * (x * y) / 2^31 |
| 166 | * |
| 167 | * For more information: https://github.com/google/gemmlowp/blob/master/public/output_stages.h#L68 |
| 168 | * |
| 169 | * In case the bias tensor is provided, the final result is: |
| 170 | * |
| 171 | * ((FixedPointMul(input[i][k] + bias[k], result_fixedpoint_multiplier)) >> result_shift) + result_offset_after_shift |
| 172 | * |
| 173 | * This function calls the following OpenCL kernels: |
| 174 | * |
| 175 | * -# @ref CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel |
| 176 | * |
| 177 | * @note The function accepts also 2 optional input arguments (min and max) which can be used to implement "rectified linear unit" activation functions |
| 178 | * after the result is shifted right by result_shift |
| 179 | */ |
| 180 | class CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPoint : public ICLSimpleFunction |
| 181 | { |
| 182 | public: |
| 183 | /** Initialise the kernel's inputs, output |
| 184 | * |
| 185 | * @param[in] input Input tensor. Data type supported: S32 |
| 186 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required. |
| 187 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 188 | * @param[out] output Output tensor. Data type supported: Data type supported: QASYMM8_SIGNED |
| 189 | * @param[in] result_fixedpoint_multiplier Fixed point value to be multiplied to each element of the input matrix when once the result_offset has been add |
| 190 | * @param[in] result_shift Number of bits to shift right the result after the fixed point multiplication |
| 191 | * @param[in] result_offset_after_shift Offset to be applied to result before converting it back to QASYMM8_SIGNED |
| 192 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8_SIGNED. Defaults to 0 |
| 193 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8_SIGNED. Defaults to 0 |
| 194 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
| 195 | */ |
| 196 | void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, |
| 197 | int min = 0, int max = 0); |
| 198 | /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPoint |
| 199 | * |
| 200 | * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
| 201 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. |
| 202 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 203 | * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8_SIGNED |
| 204 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8_SIGNED. Defaults to 0 |
| 205 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8_SIGNED. Defaults to 0 |
| 206 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
| 207 | * |
| 208 | * @return a status |
| 209 | */ |
| 210 | static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0); |
| 211 | }; |
| 212 | |
Georgios Pinitas | 51e53a3 | 2018-10-22 13:49:08 +0100 | [diff] [blame] | 213 | /** Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloat on OpenCL. |
| 214 | * |
| 215 | * This function calls the following OpenCL kernels: |
| 216 | * |
| 217 | * -# @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel |
| 218 | * |
| 219 | * @note The function accepts also 2 optional input arguments (min and max) which can be used to implement "rectified linear unit" activation functions |
| 220 | * after the result is shifted right by result_shift |
| 221 | */ |
| 222 | class CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloat : public ICLSimpleFunction |
| 223 | { |
| 224 | public: |
| 225 | /** Initialise the kernel's inputs, output |
| 226 | * |
Gian Marco Iodice | 0c54a62 | 2018-10-30 12:20:03 +0000 | [diff] [blame] | 227 | * @param[in] input Input tensor. Data type supported: S32 |
| 228 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required. |
| 229 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 230 | * @param[out] output Output tensor. Data type supported: Data type supported: QASYMM8 |
| 231 | * @param[in] multiplier Float multiplier to be multiplied to each element of the input matrix |
| 232 | * @param[in] offset Offset to be applied to result before converting it back to QASYMM8 |
| 233 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
| 234 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, |
| 235 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
Georgios Pinitas | 51e53a3 | 2018-10-22 13:49:08 +0100 | [diff] [blame] | 236 | */ |
Gian Marco Iodice | 0c54a62 | 2018-10-30 12:20:03 +0000 | [diff] [blame] | 237 | void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, float multiplier, int offset, int min = 0, int max = 0); |
Georgios Pinitas | 51e53a3 | 2018-10-22 13:49:08 +0100 | [diff] [blame] | 238 | /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint |
| 239 | * |
Gian Marco Iodice | 0c54a62 | 2018-10-30 12:20:03 +0000 | [diff] [blame] | 240 | * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
| 241 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. |
| 242 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 243 | * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8 |
| 244 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
| 245 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8, |
| 246 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions |
Georgios Pinitas | 51e53a3 | 2018-10-22 13:49:08 +0100 | [diff] [blame] | 247 | * |
| 248 | * @return a status |
| 249 | */ |
Gian Marco Iodice | 0c54a62 | 2018-10-30 12:20:03 +0000 | [diff] [blame] | 250 | static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0); |
Georgios Pinitas | 51e53a3 | 2018-10-22 13:49:08 +0100 | [diff] [blame] | 251 | }; |
Manuel Bottini | 9c9b70b | 2019-07-01 17:35:56 +0100 | [diff] [blame] | 252 | /** Basic function to execute CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint on OpenCL. |
| 253 | * |
| 254 | * CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint depends on 2 parameters: |
| 255 | * |
| 256 | * result_fixedpoint_multiplier, result_shift |
| 257 | * |
| 258 | * The final result is: |
| 259 | * |
| 260 | * (FixedPointMul(input[i][k], result_fixedpoint_multiplier) >> result_shift) |
| 261 | * |
| 262 | * where FixedPointMul(x, y) is the nearest integer to the following |
| 263 | * mathematical expression, evaluated without overflow or intermediate rounding: |
| 264 | * |
| 265 | * (x * y) / 2^31 |
| 266 | * |
| 267 | * For more information: https://github.com/google/gemmlowp/blob/master/public/output_stages.h#L68 |
| 268 | * |
| 269 | * In case the bias tensor is provided, the final result is: |
| 270 | * |
| 271 | * ((FixedPointMul(input[i][k] + bias[k], result_fixedpoint_multiplier)) >> result_shift) + result_offset_after_shift |
| 272 | * |
| 273 | * This function calls the following NEON kernels: |
| 274 | * |
| 275 | * -# @ref CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel |
| 276 | * |
| 277 | * @note The function accepts also 2 optional input arguments (min and max) which can be used to implement "rectified linear unit" activation functions |
| 278 | * after the result is shifted right by result_shift |
| 279 | */ |
| 280 | class CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint : public ICLSimpleFunction |
| 281 | { |
| 282 | public: |
| 283 | /** Initialise the kernel's inputs, output |
| 284 | * |
| 285 | * @param[in] input Input tensor. Data type supported: S32 |
| 286 | * @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required. |
| 287 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 288 | * @param[out] output Output tensor. Data type supported: Data type supported: QSYMM16 |
| 289 | * @param[in] result_fixedpoint_multiplier Fixed point value to be multiplied to each element of the input matrix when once the result_offset has been add |
| 290 | * @param[in] result_shift Number of bits to shift right the result after the fixed point multiplication |
| 291 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QSYMM16. Defaults to 0. |
| 292 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QSYMM16. |
| 293 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions. Defaults to 0. |
| 294 | */ |
| 295 | void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_fixedpoint_multiplier, int result_shift, int min = 0, int max = 0); |
| 296 | /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint |
| 297 | * |
| 298 | * @param[in] input Input tensor info. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
| 299 | * @param[in] bias Biases tensor info. Only shared biases supported and it can be a nullptr if the addition of biases is not required. |
| 300 | * Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input. |
| 301 | * @param[in] output Output tensor info. Data type supported: Data type supported: QSYMM16 |
| 302 | * @param[in] min (Optional) Min value used to saturate down the output result before converting back to QSYMM16. Defaults to 0. |
| 303 | * @param[in] max (Optional) Max value used to saturate up the output result before converting back to QSYMM16, |
| 304 | * Along with @p min, this value can be used to implement "rectified linear unit" activation functions. Defaults to 0. |
| 305 | * |
| 306 | * @return a status |
| 307 | */ |
| 308 | static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0); |
| 309 | }; |
Georgios Pinitas | 932491f | 2018-09-21 16:33:15 +0100 | [diff] [blame] | 310 | } // namespace arm_compute |
Michalis Spyrou | f464337 | 2019-11-29 16:17:13 +0000 | [diff] [blame] | 311 | #endif /*ARM_COMPUTE_CLGEMMLOWPOUTPUTSTAGE_H */ |