Function: set_tensor_const

Applicability

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product / Atlas A3 inference product	√
Atlas A2 training product / Atlas A2 inference product	√
Atlas training product	√
Atlas inference product	√
Atlas 200I/500 A2 inference product	√

Function Usage

Sets a constant input for an operator and specifies the memory address for storing the input tensor data and address length.

Prototype

C Prototype

        
             aclError aclSetTensorConst(aclTensorDesc *desc, void *dataBuffer, size_t length)

Python Function

        
             ret = acl.set_tensor_const(desc, dataBuffer, length)

Parameters

Parameter	Description
desc	Int, pointer address of the aclTensorDesc type. Call acl.create_tensor_desc to create data of the aclTensorDesc type in advance.
dataBuffer	Int, address of the memory for storing the constant input tensor data. The memory for storing the input tensor data of the operator must be determined based on the application running mode. For details about the memory allocation APIs, see Memory Management. If the app runs on the host, allocate host memory. If the app runs on the device, allocate device memory.
length	Int, length of the memory address, in bytes.

Parameter

Description

desc

Int, pointer address of the aclTensorDesc type.

Call acl.create_tensor_desc to create data of the aclTensorDesc type in advance.

dataBuffer

Int, address of the memory for storing the constant input tensor data.

The memory for storing the input tensor data of the operator must be determined based on the application running mode. For details about the memory allocation APIs, see Memory Management.

If the app runs on the host, allocate host memory.
If the app runs on the device, allocate device memory.

length

Int, length of the memory address, in bytes.

Returns

Return Value	Description
ret	Int, error code. 0 indicates success. Other values indicate failure.

Parent topic: aclmdlTensorDesc