Compute Library
18.05
|
Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint on OpenCL. More...
#include <CLGEMMLowpOutputStage.h>
Public Member Functions | |
void | configure (const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, int min=0, int max=0) |
Initialise the kernel's inputs, output. More... | |
Public Member Functions inherited from ICLSimpleFunction | |
ICLSimpleFunction () | |
Default constructor. More... | |
void | run () overridefinal |
Run the kernels contained in the function. More... | |
Public Member Functions inherited from IFunction | |
virtual | ~IFunction ()=default |
Destructor. More... | |
virtual void | prepare () |
Prepare the function for executing. More... | |
Static Public Member Functions | |
static Status | validate (const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min=0, int max=0) |
Static function to check if given info will lead to a valid configuration of CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint. More... | |
Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint on OpenCL.
CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint depends on 3 parameters:
result_fixedpoint_multiplier, result_shift, result_offset_after_shift
The final result is:
(FixedPointMul(input[i][k], result_fixedpoint_multiplier) >> result_shift) + result_offset_after_shift
where FixedPointMul(x, y) is the nearest integer to the following mathematical expression, evaluated without overflow or intermediate rounding:
(x * y) / 2^31
For more information: https://github.com/google/gemmlowp/blob/master/public/output_stages.h#L68
In case the bias tensor is provided, the final result is:
((FixedPointMul(input[i][k] + bias[k], result_fixedpoint_multiplier)) >> result_shift) + result_offset_after_shift
This function calls the following OpenCL kernels:
Definition at line 119 of file CLGEMMLowpOutputStage.h.
void configure | ( | const ICLTensor * | input, |
const ICLTensor * | bias, | ||
ICLTensor * | output, | ||
int | result_fixedpoint_multiplier, | ||
int | result_shift, | ||
int | result_offset_after_shift, | ||
int | min = 0 , |
||
int | max = 0 |
||
) |
Initialise the kernel's inputs, output.
[in] | input | Input tensor. Data type supported: S32 |
[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 input . |
[out] | output | Output tensor. Data type supported: Data type supported: QASYMM8 |
[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 |
[in] | result_shift | Number of bits to shift right the result after the fixed point multiplication |
[in] | result_offset_after_shift | Offset to be applied to result before converting it back to QASYMM8 |
[in] | min | (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
[in] | max | (Optional) Max value used to saturate up the output result before converting back to QASYMM8, Along with min , this value can be used to implement "rectified linear unit" activation functions |
Referenced by arm_compute::test::validation::DATA_TEST_CASE().
|
static |
Static function to check if given info will lead to a valid configuration of CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint.
[in] | input | Input tensor. It is the output of CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 |
[in] | bias | Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required. Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as input . |
[in] | output | Output tensor. Data type supported: Data type supported: QASYMM8 |
[in] | min | (Optional) Min value used to saturate down the output result before converting back to QASYMM8 |
[in] | max | (Optional) Max value used to saturate up the output result before converting back to QASYMM8, Along with min , this value can be used to implement "rectified linear unit" activation functions |