Function: get_dataset_tensor_desc

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	x

Function Usage

Obtains the tensor description of the specified input or output from the data of the aclmdlDataset type (used to describe the input and output data during model inference) after the model is executed if the input or output shape of a model is dynamic.

Typical scenario: If the input shape of a model is dynamic, call acl.mdl.set_dataset_tensor_desc to set the tensor description before the model is executed. After the model is executed, call acl.mdl.get_dataset_tensor_desc to obtain the tensor description of the dynamic output of the model.

Prototype

C Prototype

aclTensorDesc *aclmdlGetDatasetTensorDesc(const aclmdlDataset *dataset, size_t index)

Python Function

tensorDesc =acl.mdl.get_dataset_tensor_desc(dataset, index)

Parameters

Parameter	Description
dataset	Int, pointer address of the input or output data structure for model execution.
index	Int, input or output index. When a model has multiple inputs and outputs, call acl.mdl.get_input_name_by_index or acl.mdl.get_output_name_by_index to obtain the input or output names and set the index accordingly.

Parameter

Description

dataset

Int, pointer address of the input or output data structure for model execution.

index

Int, input or output index.

When a model has multiple inputs and outputs, call acl.mdl.get_input_name_by_index or acl.mdl.get_output_name_by_index to obtain the input or output names and set the index accordingly.

Returns

Return Value	Description
tensor_desc	Int, pointer address of the tensor description of the specified input or output.

Parent topic: aclmdlDataset