CoreFlow 1.0.0
A modern orchestration and execution runtime
Loading...
Searching...
No Matches
Kernel: Tensor Matrix Multiply

Classes

struct  _vx_tensor_matrix_multiply_params_t
 Matrix Multiply Parameters. More...
 

Typedefs

typedef struct _vx_tensor_matrix_multiply_params_t vx_tensor_matrix_multiply_params_t
 Matrix Multiply Parameters.
 

Functions

VX_API_ENTRY vx_node VX_API_CALL vxTensorMatrixMultiplyNode (vx_graph graph, vx_tensor input1, vx_tensor input2, vx_tensor input3, const vx_tensor_matrix_multiply_params_t *matrix_multiply_params, vx_tensor output)
 [Graph] Creates a generalized matrix multiplication node.
 
VX_API_ENTRY vx_status VX_API_CALL vxuTensorMatrixMultiply (vx_context context, vx_tensor input1, vx_tensor input2, vx_tensor input3, const vx_tensor_matrix_multiply_params_t *matrix_multiply_params, vx_tensor output)
 [Immediate] Performs a generalized matrix multiplication.
 

Detailed Description

Typedef Documentation

◆ vx_tensor_matrix_multiply_params_t

#include <vx_types.h>

Matrix Multiply Parameters.

transpose_input1/input2/input3 : if True the matrix is transposed before the operation, otherwise the matrix is used as is.

Function Documentation

◆ vxTensorMatrixMultiplyNode()

VX_API_ENTRY vx_node VX_API_CALL vxTensorMatrixMultiplyNode ( vx_graph graph,
vx_tensor input1,
vx_tensor input2,
vx_tensor input3,
const vx_tensor_matrix_multiply_params_t * matrix_multiply_params,
vx_tensor output )

#include <vx_nodes.h>

[Graph] Creates a generalized matrix multiplication node.

Parameters
[in]graphThe reference to the graph.
[in]input1The first input 2D tensor of type VX_TYPE_INT16 with fixed_point_pos 8, or tensor data types VX_TYPE_UINT8 or VX_TYPE_INT8, with fixed_point_pos 0.
[in]input2The second 2D tensor. Must be in the same data type as input1.
[in]input3The third 2D tensor. Must be in the same data type as input1. [optional].
[in]matrix_multiply_paramsMatrix multiply parameters, see vx_tensor_matrix_multiply_params_t .
[out]outputThe output 2D tensor. Must be in the same data type as input1. Output dimension must agree the formula in the description.
Returns
vx_node.
A node reference vx_node. Any possible errors preventing a successful creation should be checked using vxGetStatus.

◆ vxuTensorMatrixMultiply()

VX_API_ENTRY vx_status VX_API_CALL vxuTensorMatrixMultiply ( vx_context context,
vx_tensor input1,
vx_tensor input2,
vx_tensor input3,
const vx_tensor_matrix_multiply_params_t * matrix_multiply_params,
vx_tensor output )

#include <vxu.h>

[Immediate] Performs a generalized matrix multiplication.

Parameters
[in]contextThe reference to the overall context.
[in]input1The first input 2D tensor of type VX_TYPE_INT16 with fixed_point_pos 8, or tensor data types VX_TYPE_UINT8 or VX_TYPE_INT8, with fixed_point_pos 0.
[in]input2The second 2D tensor. Must be in the same data type as input1.
[in]input3The third 2D tensor. Must be in the same data type as input1. [optional].
[in]matrix_multiply_paramsMatrix multiply parameters, see vx_tensor_matrix_multiply_params_t .
[out]outputThe output 2D tensor. Must be in the same data type as input1. Output dimension must agree the formula in the description.
Returns
A vx_status_e enumeration.
Return values
VX_SUCCESSSuccess
*An error occurred. See vx_status_e.