Other functions need to quantization

These function are used to accelerate quantization algorithms in Synet Framework. More...

Functions
SIMD_API void	SimdSynetDequantizeLinear (const uint8_t *src, size_t size, int32_t bias, const float *norm, float *dst)
	Performs UINT8 linear dequantization. More...
SIMD_API void	SimdSynetQuantizedConcatLayerForward (size_t count, const uint8_t **src, size_t num, const size_t *size, const int32_t *bias, const float *norm, const float *scale, int32_t zero, uint8_t *dst)
	This function is used for forward propagation of QuantizedConcatLayer. More...
SIMD_API void	SimdSynetQuantizedShuffleLayerForward (const uint8_t *src0, int bias0, const float *norm0, size_t srcC0, const uint8_t *src1, int bias1, const float *norm1, size_t srcC1, size_t spatial, uint8_t *dst0, uint8_t *dst1, const float *scale, int zero, SimdTensorFormatType format, int type)
	This function is used for forward propagation of QuantizedShuffleLayer. More...
SIMD_API void	SimdSynetQuantizeLinear (const float *src, size_t size, const float *norm, int32_t zero, uint8_t *dst)
	Performs UINT8 linear quantization. More...

Detailed Description

These function are used to accelerate quantization algorithms in Synet Framework.

Function Documentation

SimdSynetDequantizeLinear()

void SimdSynetDequantizeLinear	(	const uint8_t *	src,
		size_t	size,
		int32_t	bias,
		const float *	norm,
		float *	dst
	)

Performs UINT8 linear dequantization.

Algorithm's details for SimdSynetDequantizeLinear:

for(i = 0; i < size; ++i)
    dst[i] = (src[i] + bias) * norm[0];

Parameters

[in]	src	- a pointer to UINT8 input tensor.
[in]	size	- a size of the input and output tensors.
[in]	bias	- a dequantization bias (-zero).
[in]	norm	- a dequantization norm (scale).
[out]	dst	- a pointer to FP32 output tensor.

SimdSynetQuantizedConcatLayerForward()

void SimdSynetQuantizedConcatLayerForward	(	size_t	count,
		const uint8_t **	src,
		size_t	num,
		const size_t *	size,
		const int32_t *	bias,
		const float *	norm,
		const float *	scale,
		int32_t	zero,
		uint8_t *	dst
	)

This function is used for forward propagation of QuantizedConcatLayer.

Note

This function is used in Synet Framework.

Parameters

[in]	count	- a number of input tensors.
[in]	src	- an array with pointers to UINT8 input tensors.
[in]	num	- an output size of input/output tensors (number of concatenations).
[in]	size	- an array with concatenation sizes of input tensors.
[in]	bias	- an array with bias parameter of input tensors (-zero).
[in]	norm	- an array with dequantization norm parameter of input tensors (scale).
[in]	scale	- an output quantization norm (1/scale).
[in]	zero	- an output quantization zero.
[out]	dst	- a pointer to the UINT8 output tensor.

SimdSynetQuantizedShuffleLayerForward()

void SimdSynetQuantizedShuffleLayerForward	(	const uint8_t *	src0,
		int	bias0,
		const float *	norm0,
		size_t	srcC0,
		const uint8_t *	src1,
		int	bias1,
		const float *	norm1,
		size_t	srcC1,
		size_t	spatial,
		uint8_t *	dst0,
		uint8_t *	dst1,
		const float *	scale,
		int	zero,
		SimdTensorFormatType	format,
		int	type
	)

This function is used for forward propagation of QuantizedShuffleLayer.

Note

This function is used in Synet Framework.

Parameters

[in]	src0	- a pointer to the 8-bit integer array with the first input tensor.
[in]	bias0	- a dequantization bias parameter of the first input tensor (-zero).
[in]	norm0	- a dequantization norm parameter of the first input tensor (scale).
[in]	srcC0	- a number of channels in the first input tensor.
[in]	src1	- a pointer to the 8-bit integer array with the second input tensor.
[in]	bias1	- a dequantization bias parameter of the second input tensor (-zero).
[in]	norm1	- a dequantization norm parameter of the second input tensor (scale).
[in]	srcC1	- a number of channels in the second input tensor.
[in]	spatial	- a spatial size of (input/output) tensors.
[out]	dst0	- a pointer to the 8-bit integer array with the first output tensor.
[out]	dst1	- a pointer to the 8-bit integer array with the second output tensor.
[in]	scale	- an output quantization norm (1/scale).
[in]	zero	- an output quantization zero.
[in]	format	- a format of (input/output) tensors.
[in]	type	- a shuffle type (it can be 0 or 1).

SimdSynetQuantizeLinear()

void SimdSynetQuantizeLinear	(	const float *	src,
		size_t	size,
		const float *	norm,
		int32_t	zero,
		uint8_t *	dst
	)

Performs UINT8 linear quantization.

Algorithm's details for SimdSynetQuantizeLinear:

for(i = 0; i < size; ++i)
    dst[i] = Min(Max(std::nearbyint(src[i] * scale[0]) + zero), 0), 255);

Parameters

[in]	src	- a pointer to FP32 input tensor.
[in]	size	- a size of the input and output tensors.
[in]	norm	- a quantization norm (1/scale).
[in]	zero	- a quantization zero.
[out]	dst	- a pointer to UINT8 output tensor.