dpuGetInputTensorScale()

dpuGetInputTensorScale() - 1.2 English

Vitis AI User Guide (UG1414)

Document ID

UG1414

Release Date

2020-07-21

Version

1.2 English

Synopsis

float dpuGetInputTensorScale
(
DPUTask *task,
const char*nodeName,
int idx = 0
    );

Arguments

task: The pointer to DPU Task.
nodeName: The pointer to DPU Node’s name.
Note: the available names of one DPU Kernel’s or Task’s output Node are listed out after a neural network is compiled by VAI_C. If invalid Node name specified, failure message is reported.
idx: The index of a single output tensor for the Node, with default value as 0.

Description

Get the scale value of DPU Task’s input Tensor. For each DPU input Tensor, it has one unified scale value indicating its quantization information for reformatting between data types of INT8 and FP32.

Returns

The scale value of Task’s input Tensor on success, or report error in case of any failure.

Include File

n2cube.h

Availability