BatchVector

Function Usage

Combine batches into a tensor. This function calls BatchConcat and BatchStack to perform batch combination. The keepDims parameter is used to control the function to be called.

Prototype

static APP_ERROR BatchVector(const std::vector<TensorBase> &inputs, TensorBase &output, const bool &keepDims = false);

Parameter Description

Parameter	Input/Output	Description
inputs	Input	For a multi-batch model, the TensorBase object list is used as the output. The list size cannot exceed 1024.
output	Output	TensorBase object after the batches are combined.
keepDims	Input	Parameter that retains the multidimensional feature. true indicates that BatchConcat is called to retain the multidimensional feature. false indicates that BatchStack is not called to retain the multidimensional feature.

Parameter

Input/Output

Description

inputs

Input

For a multi-batch model, the TensorBase object list is used as the output. The list size cannot exceed 1024.

output

Output

TensorBase object after the batches are combined.

keepDims

Input

Parameter that retains the multidimensional feature.

true indicates that BatchConcat is called to retain the multidimensional feature.
false indicates that BatchStack is not called to retain the multidimensional feature.

Return Parameter Description

Data Structure	Description
APP_ERROR	Error code returned during program execution. For details, see the MxBase/ErrorCode/ErrorCode.h file.

Parent topic: TensorBase