Fully Connected

Functions
arm_cmsis_nn_status	arm_nn_vec_mat_mul_result_acc_s16 (const int16_t *lhs, const int8_t *rhs, const int64_t *effective_bias, int16_t *dst, const int32_t dst_multiplier, const int32_t dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t batches, const int32_t batch_offset)
	The result of the multiplication is accumulated to the passed result buffer. Multiplies a matrix by a "batched" vector (i.e. a matrix with a batch dimension composed by input vectors independent from each other).
arm_cmsis_nn_status	arm_nn_vec_mat_mult_t_per_ch_s8 (const int8_t *lhs, const int8_t *rhs, const int32_t *kernel_sum, const int32_t *bias, int8_t *dst, const int32_t lhs_offset, const int32_t dst_offset, const int32_t *dst_multiplier, const int32_t *dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t activation_min, const int32_t activation_max, const int32_t address_offset, const int32_t rhs_offset)
	s8 Vector by Matrix (transposed) multiplication using per channel quantization for output
arm_cmsis_nn_status	arm_nn_vec_mat_mult_t_s16 (const int16_t *lhs, const int8_t *rhs, const int64_t *bias, int16_t *dst, const int32_t dst_multiplier, const int32_t dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t activation_min, const int32_t activation_max)
	s16 Vector by s8 Matrix (transposed) multiplication
arm_cmsis_nn_status	arm_nn_vec_mat_mult_t_s16_s16 (const int16_t *lhs, const int16_t *rhs, const int64_t *bias, int16_t *dst, const int32_t dst_multiplier, const int32_t dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t activation_min, const int32_t activation_max)
	s16 Vector by s16 Matrix (transposed) multiplication
arm_cmsis_nn_status	arm_nn_vec_mat_mult_t_s4 (const int8_t *lhs, const int8_t *packed_rhs, const int32_t *bias, int8_t *dst, const int32_t lhs_offset, const int32_t dst_offset, const int32_t dst_multiplier, const int32_t dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t activation_min, const int32_t activation_max)
	s4 Vector by Matrix (transposed) multiplication
arm_cmsis_nn_status	arm_nn_vec_mat_mult_t_s8 (const int8_t *lhs, const int8_t *rhs, const int32_t *kernel_sum, const int32_t *bias, int8_t *dst, const int32_t lhs_offset, const int32_t dst_offset, const int32_t dst_multiplier, const int32_t dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t activation_min, const int32_t activation_max, const int32_t address_offset, const int32_t rhs_offset)
	s8 Vector by Matrix (transposed) multiplication
arm_cmsis_nn_status	arm_nn_vec_mat_mult_t_svdf_s8 (const int8_t *lhs, const int8_t *rhs, int16_t *dst, const int32_t lhs_offset, const int32_t dst_offset, const int32_t dst_multiplier, const int32_t dst_shift, const int32_t rhs_cols, const int32_t rhs_rows, const int32_t activation_min, const int32_t activation_max)
	s8 Vector by Matrix (transposed) multiplication with s16 output

Description

Support functions for Fully Connected

Function Documentation

arm_nn_vec_mat_mul_result_acc_s16()

arm_cmsis_nn_status arm_nn_vec_mat_mul_result_acc_s16	(	const int16_t *	lhs,
		const int8_t *	rhs,
		const int64_t *	effective_bias,
		int16_t *	dst,
		const int32_t	dst_multiplier,
		const int32_t	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	batches,
		const int32_t	batch_offset
	)

The result of the multiplication is accumulated to the passed result buffer. Multiplies a matrix by a "batched" vector (i.e. a matrix with a batch dimension composed by input vectors independent from each other).

Parameters

[in]	lhs	Batched vector
[in]	rhs	Weights - input matrix (H(Rows)xW(Columns))
[in]	effective_bias	Bias + lhs_offset * kernel_sum term precalculated into a constant vector.
[out]	dst	Output
[in]	dst_multiplier	Multiplier for quantization
[in]	dst_shift	Shift for quantization
[in]	rhs_cols	Vector/matarix column length
[in]	rhs_rows	Row count of matrix
[in]	batches	Batch size
[in]	batch_offset	Number of timesteps between consecutive batches in input, see arm_nn_lstm_step_s16. Note that the output is always stored with sequential batches.

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_nn_vec_mat_mult_t_per_ch_s8()

arm_cmsis_nn_status arm_nn_vec_mat_mult_t_per_ch_s8	(	const int8_t *	lhs,
		const int8_t *	rhs,
		const int32_t *	kernel_sum,
		const int32_t *	bias,
		int8_t *	dst,
		const int32_t	lhs_offset,
		const int32_t	dst_offset,
		const int32_t *	dst_multiplier,
		const int32_t *	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	activation_min,
		const int32_t	activation_max,
		const int32_t	address_offset,
		const int32_t	rhs_offset
	)

s8 Vector by Matrix (transposed) multiplication using per channel quantization for output

Parameters

[in]	lhs	Input left-hand side vector
[in]	rhs	Input right-hand side matrix (transposed)
[in]	kernel_sum	Kernel sums of the kernels (rhs). See arm_vector_sum_s8 for more info.
[in]	bias	Input bias
[out]	dst	Output vector
[in]	lhs_offset	Offset to be added to the input values of the left-hand side vector. Range: -127 to 128
[in]	dst_offset	Offset to be added to the output values. Range: -127 to 128
[in]	dst_multiplier	Output multipliers
[in]	dst_shift	Output shifts
[in]	rhs_cols	Number of columns in the right-hand side input matrix
[in]	rhs_rows	Number of rows in the right-hand side input matrix
[in]	activation_min	Minimum value to clamp the output to. Range: int8
[in]	activation_max	Maximum value to clamp the output to. Range: int8
[in]	address_offset	Memory position offset for dst. First output is stored at 'dst', the second at 'dst + address_offset' and so on. Default value is typically 1.
[in]	rhs_offset	Offset to be added to the input values of the right-hand side vector. Range: -127 to 128

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_nn_vec_mat_mult_t_s16()

arm_cmsis_nn_status arm_nn_vec_mat_mult_t_s16	(	const int16_t *	lhs,
		const int8_t *	rhs,
		const int64_t *	bias,
		int16_t *	dst,
		const int32_t	dst_multiplier,
		const int32_t	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	activation_min,
		const int32_t	activation_max
	)

s16 Vector by s8 Matrix (transposed) multiplication

Parameters

[in]	lhs	Input left-hand side vector
[in]	rhs	Input right-hand side matrix (transposed)
[in]	bias	Input bias
[out]	dst	Output vector
[in]	dst_multiplier	Output multiplier
[in]	dst_shift	Output shift
[in]	rhs_cols	Number of columns in the right-hand side input matrix
[in]	rhs_rows	Number of rows in the right-hand side input matrix
[in]	activation_min	Minimum value to clamp the output to. Range: int16
[in]	activation_max	Maximum value to clamp the output to. Range: int16

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_nn_vec_mat_mult_t_s16_s16()

arm_cmsis_nn_status arm_nn_vec_mat_mult_t_s16_s16	(	const int16_t *	lhs,
		const int16_t *	rhs,
		const int64_t *	bias,
		int16_t *	dst,
		const int32_t	dst_multiplier,
		const int32_t	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	activation_min,
		const int32_t	activation_max
	)

s16 Vector by s16 Matrix (transposed) multiplication

Parameters

[in]	lhs	Input left-hand side vector
[in]	rhs	Input right-hand side matrix (transposed)
[in]	bias	Input bias
[out]	dst	Output vector
[in]	dst_multiplier	Output multiplier
[in]	dst_shift	Output shift
[in]	rhs_cols	Number of columns in the right-hand side input matrix
[in]	rhs_rows	Number of rows in the right-hand side input matrix
[in]	activation_min	Minimum value to clamp the output to. Range: int16
[in]	activation_max	Maximum value to clamp the output to. Range: int16

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_nn_vec_mat_mult_t_s4()

arm_cmsis_nn_status arm_nn_vec_mat_mult_t_s4	(	const int8_t *	lhs,
		const int8_t *	packed_rhs,
		const int32_t *	bias,
		int8_t *	dst,
		const int32_t	lhs_offset,
		const int32_t	dst_offset,
		const int32_t	dst_multiplier,
		const int32_t	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	activation_min,
		const int32_t	activation_max
	)

s4 Vector by Matrix (transposed) multiplication

Parameters

[in]	lhs	Input left-hand side vector
[in]	packed_rhs	Input right-hand side matrix (transposed)
[in]	bias	Input bias
[out]	dst	Output vector
[in]	lhs_offset	Offset to be added to the input values of the left-hand side vector. Range: -127 to 128
[in]	dst_offset	Offset to be added to the output values. Range: -127 to 128
[in]	dst_multiplier	Output multiplier
[in]	dst_shift	Output shift
[in]	rhs_cols	Number of columns in the right-hand side input matrix
[in]	rhs_rows	Number of rows in the right-hand side input matrix
[in]	activation_min	Minimum value to clamp the output to. Range: int8
[in]	activation_max	Maximum value to clamp the output to. Range: int8

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_nn_vec_mat_mult_t_s8()

arm_cmsis_nn_status arm_nn_vec_mat_mult_t_s8	(	const int8_t *	lhs,
		const int8_t *	rhs,
		const int32_t *	kernel_sum,
		const int32_t *	bias,
		int8_t *	dst,
		const int32_t	lhs_offset,
		const int32_t	dst_offset,
		const int32_t	dst_multiplier,
		const int32_t	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	activation_min,
		const int32_t	activation_max,
		const int32_t	address_offset,
		const int32_t	rhs_offset
	)

s8 Vector by Matrix (transposed) multiplication

Parameters

[in]	lhs	Input left-hand side vector
[in]	rhs	Input right-hand side matrix (transposed)
[in]	kernel_sum	Kernel sums of the kernels (rhs). See arm_vector_sum_s8 for more info.
[in]	bias	Input bias
[out]	dst	Output vector
[in]	lhs_offset	Offset to be added to the input values of the left-hand side vector. Range: -127 to 128
[in]	dst_offset	Offset to be added to the output values. Range: -127 to 128
[in]	dst_multiplier	Output multiplier
[in]	dst_shift	Output shift
[in]	rhs_cols	Number of columns in the right-hand side input matrix
[in]	rhs_rows	Number of rows in the right-hand side input matrix
[in]	activation_min	Minimum value to clamp the output to. Range: int8
[in]	activation_max	Maximum value to clamp the output to. Range: int8
[in]	address_offset	Memory position offset for dst. First output is stored at 'dst', the second at 'dst + address_offset' and so on. Default value is typically 1.
[in]	rhs_offset	Offset to be added to the input values of the right-hand side vector. Range: -127 to 128

Returns

The function returns ARM_CMSIS_NN_SUCCESS

arm_nn_vec_mat_mult_t_svdf_s8()

arm_cmsis_nn_status arm_nn_vec_mat_mult_t_svdf_s8	(	const int8_t *	lhs,
		const int8_t *	rhs,
		int16_t *	dst,
		const int32_t	lhs_offset,
		const int32_t	scatter_offset,
		const int32_t	dst_multiplier,
		const int32_t	dst_shift,
		const int32_t	rhs_cols,
		const int32_t	rhs_rows,
		const int32_t	activation_min,
		const int32_t	activation_max
	)

s8 Vector by Matrix (transposed) multiplication with s16 output

Parameters

[in]	lhs	Input left-hand side vector
[in]	rhs	Input right-hand side matrix (transposed)
[out]	dst	Output vector
[in]	lhs_offset	Offset to be added to the input values of the left-hand side vector. Range: -127 to 128
[in]	scatter_offset	Address offset for dst. First output is stored at 'dst', the second at 'dst + scatter_offset' and so on.
[in]	dst_multiplier	Output multiplier
[in]	dst_shift	Output shift
[in]	rhs_cols	Number of columns in the right-hand side input matrix
[in]	rhs_rows	Number of rows in the right-hand side input matrix
[in]	activation_min	Minimum value to clamp the output to. Range: int16
[in]	activation_max	Maximum value to clamp the output to. Range: int16

Returns

The function returns ARM_CMSIS_NN_SUCCESS