Elementwise Functions

Functions
arm_cmsis_nn_status	arm_elementwise_add_f32 (const float32_t *input_1_vect, const float32_t *input_2_vect, float32_t *output, float32_t out_activation_min, float32_t out_activation_max, int32_t block_size)
	Elementwise add with optional output clamp.
arm_cmsis_nn_status	arm_elementwise_mul_f32 (const float32_t *input_1_vect, const float32_t *input_2_vect, float32_t *output, float32_t out_activation_min, float32_t out_activation_max, int32_t block_size)
	Elementwise multiply with optional output clamp.
arm_cmsis_nn_status	arm_minimum_f32 (const cmsis_nn_context *ctx, const float32_t *input_1_data, const cmsis_nn_dims *input_1_dims, const float32_t *input_2_data, const cmsis_nn_dims *input_2_dims, float32_t *output_data, const cmsis_nn_dims *output_dims)
	Elementwise minimum.
arm_cmsis_nn_status	arm_maximum_f32 (const cmsis_nn_context *ctx, const float32_t *input_1_data, const cmsis_nn_dims *input_1_dims, const float32_t *input_2_data, const cmsis_nn_dims *input_2_dims, float32_t *output_data, const cmsis_nn_dims *output_dims)
	Elementwise maximum.
arm_cmsis_nn_status	arm_elementwise_add_f16 (const float16_t *input_1_vect, const float16_t *input_2_vect, float16_t *output, float16_t out_activation_min, float16_t out_activation_max, int32_t block_size)
	Elementwise add with optional output clamp.
arm_cmsis_nn_status	arm_elementwise_mul_f16 (const float16_t *input_1_vect, const float16_t *input_2_vect, float16_t *output, float16_t out_activation_min, float16_t out_activation_max, int32_t block_size)
	Elementwise multiply with optional output clamp.
arm_cmsis_nn_status	arm_minimum_f16 (const cmsis_nn_context *ctx, const float16_t *input_1_data, const cmsis_nn_dims *input_1_dims, const float16_t *input_2_data, const cmsis_nn_dims *input_2_dims, float16_t *output_data, const cmsis_nn_dims *output_dims)
	Elementwise minimum.
arm_cmsis_nn_status	arm_maximum_f16 (const cmsis_nn_context *ctx, const float16_t *input_1_data, const cmsis_nn_dims *input_1_dims, const float16_t *input_2_data, const cmsis_nn_dims *input_2_dims, float16_t *output_data, const cmsis_nn_dims *output_dims)
	Elementwise maximum.
arm_cmsis_nn_status	arm_elementwise_add_s16 (const int16_t *input_1_vect, const int16_t *input_2_vect, const int32_t input_1_offset, const int32_t input_1_mult, const int32_t input_1_shift, const int32_t input_2_offset, const int32_t input_2_mult, const int32_t input_2_shift, const int32_t left_shift, int16_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size)
	s16 elementwise add of two vectors
arm_cmsis_nn_status	arm_elementwise_add_s8 (const int8_t *input_1_vect, const int8_t *input_2_vect, const int32_t input_1_offset, const int32_t input_1_mult, const int32_t input_1_shift, const int32_t input_2_offset, const int32_t input_2_mult, const int32_t input_2_shift, const int32_t left_shift, int8_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size)
	s8 elementwise add of two vectors
arm_cmsis_nn_status	arm_elementwise_mul_acc_s16 (const int16_t *input_1_vect, const int16_t *input_2_vect, const int32_t input_1_offset, const int32_t input_2_offset, int16_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size)
	s16 element wise accumulative multiplication of two vectors
arm_cmsis_nn_status	arm_elementwise_mul_s16 (const int16_t *input_1_vect, const int16_t *input_2_vect, const int32_t input_1_offset, const int32_t input_2_offset, int16_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size)
	s16 element wise multiplication of two vectors
arm_cmsis_nn_status	arm_elementwise_mul_s16_batch_offset (const int16_t *input_1_vect, const int16_t *input_2_vect, int16_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t block_size, const int32_t batch_size, const int32_t batch_offset)
	s16 element wise multiplication of batches of two vectors
arm_cmsis_nn_status	arm_elementwise_mul_s8 (const int8_t *input_1_vect, const int8_t *input_2_vect, const int32_t input_1_offset, const int32_t input_2_offset, int8_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size)
	s8 elementwise multiplication

Description

Elementwise add and multiplication functions.

Function Documentation

arm_elementwise_add_f16()

arm_cmsis_nn_status arm_elementwise_add_f16	(	const float16_t *	input_1_vect,
		const float16_t *	input_2_vect,
		float16_t *	output,
		float16_t	out_activation_min,
		float16_t	out_activation_max,
		int32_t	block_size
	)

Elementwise add with optional output clamp.

Parameters

[in]	input_1_vect	Pointer to the first input vector.
[in]	input_2_vect	Pointer to the second input vector.
[out]	output	Pointer to the output vector.
[in]	out_activation_min	Minimum output clamp value.
[in]	out_activation_max	Maximum output clamp value.
[in]	block_size	Number of elements to process.

Returns

ARM_CMSIS_NN_SUCCESS on success or ARM_CMSIS_NN_ARG_ERROR on invalid arguments.

arm_elementwise_add_f32()

arm_cmsis_nn_status arm_elementwise_add_f32	(	const float32_t *	input_1_vect,
		const float32_t *	input_2_vect,
		float32_t *	output,
		float32_t	out_activation_min,
		float32_t	out_activation_max,
		int32_t	block_size
	)

Elementwise add with optional output clamp.

Parameters

[in]	input_1_vect	Pointer to the first input vector.
[in]	input_2_vect	Pointer to the second input vector.
[out]	output	Pointer to the output vector.
[in]	out_activation_min	Minimum output clamp value.
[in]	out_activation_max	Maximum output clamp value.
[in]	block_size	Number of elements to process.

Returns

ARM_CMSIS_NN_SUCCESS on success or ARM_CMSIS_NN_ARG_ERROR on invalid arguments.

arm_elementwise_add_s16()

arm_cmsis_nn_status arm_elementwise_add_s16	(	const int16_t *	input_1_vect,
		const int16_t *	input_2_vect,
		const int32_t	input_1_offset,
		const int32_t	input_1_mult,
		const int32_t	input_1_shift,
		const int32_t	input_2_offset,
		const int32_t	input_2_mult,
		const int32_t	input_2_shift,
		const int32_t	left_shift,
		int16_t *	output,
		const int32_t	out_offset,
		const int32_t	out_mult,
		const int32_t	out_shift,
		const int32_t	out_activation_min,
		const int32_t	out_activation_max,
		const int32_t	block_size
	)

s16 elementwise add of two vectors

Parameters

[in]	input_1_vect	pointer to input vector 1
[in]	input_2_vect	pointer to input vector 2
[in]	input_1_offset	offset for input 1. Not used.
[in]	input_1_mult	multiplier for input 1
[in]	input_1_shift	shift for input 1
[in]	input_2_offset	offset for input 2. Not used.
[in]	input_2_mult	multiplier for input 2
[in]	input_2_shift	shift for input 2
[in]	left_shift	input left shift
[in,out]	output	pointer to output vector
[in]	out_offset	output offset. Not used.
[in]	out_mult	output multiplier
[in]	out_shift	output shift
[in]	out_activation_min	minimum value to clamp output to. Min: -32768
[in]	out_activation_max	maximum value to clamp output to. Max: 32767
[in]	block_size	number of samples

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_elementwise_add_s8()

arm_cmsis_nn_status arm_elementwise_add_s8	(	const int8_t *	input_1_vect,
		const int8_t *	input_2_vect,
		const int32_t	input_1_offset,
		const int32_t	input_1_mult,
		const int32_t	input_1_shift,
		const int32_t	input_2_offset,
		const int32_t	input_2_mult,
		const int32_t	input_2_shift,
		const int32_t	left_shift,
		int8_t *	output,
		const int32_t	out_offset,
		const int32_t	out_mult,
		const int32_t	out_shift,
		const int32_t	out_activation_min,
		const int32_t	out_activation_max,
		const int32_t	block_size
	)

s8 elementwise add of two vectors

Parameters

[in]	input_1_vect	pointer to input vector 1
[in]	input_2_vect	pointer to input vector 2
[in]	input_1_offset	offset for input 1. Range: -127 to 128
[in]	input_1_mult	multiplier for input 1
[in]	input_1_shift	shift for input 1
[in]	input_2_offset	offset for input 2. Range: -127 to 128
[in]	input_2_mult	multiplier for input 2
[in]	input_2_shift	shift for input 2
[in]	left_shift	input left shift
[in,out]	output	pointer to output vector
[in]	out_offset	output offset. Range: -128 to 127
[in]	out_mult	output multiplier
[in]	out_shift	output shift
[in]	out_activation_min	minimum value to clamp output to. Min: -128
[in]	out_activation_max	maximum value to clamp output to. Max: 127
[in]	block_size	number of samples

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_elementwise_mul_acc_s16()

arm_cmsis_nn_status arm_elementwise_mul_acc_s16	(	const int16_t *	input_1_vect,
		const int16_t *	input_2_vect,
		const int32_t	input_1_offset,
		const int32_t	input_2_offset,
		int16_t *	output,
		const int32_t	out_offset,
		const int32_t	out_mult,
		const int32_t	out_shift,
		const int32_t	out_activation_min,
		const int32_t	out_activation_max,
		const int32_t	block_size
	)

s16 element wise accumulative multiplication of two vectors

s16 elementwise multiplication. The result of the multiplication is accumulated to the passed result buffer.

Note

Refer header file for details.

arm_elementwise_mul_f16()

arm_cmsis_nn_status arm_elementwise_mul_f16	(	const float16_t *	input_1_vect,
		const float16_t *	input_2_vect,
		float16_t *	output,
		float16_t	out_activation_min,
		float16_t	out_activation_max,
		int32_t	block_size
	)

Elementwise multiply with optional output clamp.

Parameters

[in]	input_1_vect	Pointer to the first input vector.
[in]	input_2_vect	Pointer to the second input vector.
[out]	output	Pointer to the output vector.
[in]	out_activation_min	Minimum output clamp value.
[in]	out_activation_max	Maximum output clamp value.
[in]	block_size	Number of elements to process.

Returns

ARM_CMSIS_NN_SUCCESS on success or ARM_CMSIS_NN_ARG_ERROR on invalid arguments.

arm_elementwise_mul_f32()

arm_cmsis_nn_status arm_elementwise_mul_f32	(	const float32_t *	input_1_vect,
		const float32_t *	input_2_vect,
		float32_t *	output,
		float32_t	out_activation_min,
		float32_t	out_activation_max,
		int32_t	block_size
	)

Elementwise multiply with optional output clamp.

Parameters

[in]	input_1_vect	Pointer to the first input vector.
[in]	input_2_vect	Pointer to the second input vector.
[out]	output	Pointer to the output vector.
[in]	out_activation_min	Minimum output clamp value.
[in]	out_activation_max	Maximum output clamp value.
[in]	block_size	Number of elements to process.

Returns

ARM_CMSIS_NN_SUCCESS on success or ARM_CMSIS_NN_ARG_ERROR on invalid arguments.

arm_elementwise_mul_s16()

arm_cmsis_nn_status arm_elementwise_mul_s16	(	const int16_t *	input_1_vect,
		const int16_t *	input_2_vect,
		const int32_t	input_1_offset,
		const int32_t	input_2_offset,
		int16_t *	output,
		const int32_t	out_offset,
		const int32_t	out_mult,
		const int32_t	out_shift,
		const int32_t	out_activation_min,
		const int32_t	out_activation_max,
		const int32_t	block_size
	)

s16 element wise multiplication of two vectors

s16 elementwise multiplication

Note

Refer header file for details.

arm_elementwise_mul_s16_batch_offset()

arm_cmsis_nn_status arm_elementwise_mul_s16_batch_offset	(	const int16_t *	input_1_vect,
		const int16_t *	input_2_vect,
		int16_t *	output,
		const int32_t	out_offset,
		const int32_t	out_mult,
		const int32_t	out_shift,
		const int32_t	block_size,
		const int32_t	batch_size,
		const int32_t	batch_offset
	)

s16 element wise multiplication of batches of two vectors

s16 elementwise multiplication with s16 output

Note

Refer header file for details.

arm_elementwise_mul_s8()

arm_cmsis_nn_status arm_elementwise_mul_s8	(	const int8_t *	input_1_vect,
		const int8_t *	input_2_vect,
		const int32_t	input_1_offset,
		const int32_t	input_2_offset,
		int8_t *	output,
		const int32_t	out_offset,
		const int32_t	out_mult,
		const int32_t	out_shift,
		const int32_t	out_activation_min,
		const int32_t	out_activation_max,
		const int32_t	block_size
	)

s8 elementwise multiplication

Parameters

[in]	input_1_vect	pointer to input vector 1
[in]	input_2_vect	pointer to input vector 2
[in]	input_1_offset	offset for input 1. Range: -127 to 128
[in]	input_2_offset	offset for input 2. Range: -127 to 128
[in,out]	output	pointer to output vector
[in]	out_offset	output offset. Range: -128 to 127
[in]	out_mult	output multiplier
[in]	out_shift	output shift
[in]	out_activation_min	minimum value to clamp output to. Min: -128
[in]	out_activation_max	maximum value to clamp output to. Max: 127
[in]	block_size	number of samples

Returns

The function returns ARM_CMSIS_NN_SUCCESS

Supported framework: TensorFlow Lite micro

arm_maximum_f16()

arm_cmsis_nn_status arm_maximum_f16	(	const cmsis_nn_context *	ctx,
		const float16_t *	input_1_data,
		const cmsis_nn_dims *	input_1_dims,
		const float16_t *	input_2_data,
		const cmsis_nn_dims *	input_2_dims,
		float16_t *	output_data,
		const cmsis_nn_dims *	output_dims
	)

Elementwise maximum.

arm_maximum_f32()

arm_cmsis_nn_status arm_maximum_f32	(	const cmsis_nn_context *	ctx,
		const float32_t *	input_1_data,
		const cmsis_nn_dims *	input_1_dims,
		const float32_t *	input_2_data,
		const cmsis_nn_dims *	input_2_dims,
		float32_t *	output_data,
		const cmsis_nn_dims *	output_dims
	)

Elementwise maximum.

arm_minimum_f16()

arm_cmsis_nn_status arm_minimum_f16	(	const cmsis_nn_context *	ctx,
		const float16_t *	input_1_data,
		const cmsis_nn_dims *	input_1_dims,
		const float16_t *	input_2_data,
		const cmsis_nn_dims *	input_2_dims,
		float16_t *	output_data,
		const cmsis_nn_dims *	output_dims
	)

Elementwise minimum.

arm_minimum_f32()

arm_cmsis_nn_status arm_minimum_f32	(	const cmsis_nn_context *	ctx,
		const float32_t *	input_1_data,
		const cmsis_nn_dims *	input_1_dims,
		const float32_t *	input_2_data,
		const cmsis_nn_dims *	input_2_dims,
		float32_t *	output_data,
		const cmsis_nn_dims *	output_dims
	)

Elementwise minimum.