Class SingleShot

Definition

Namespace:
Tizen.MachineLearning.Inference
Assembly:
Tizen.MachineLearning.Inference.dll
API Level:
6

The SingleShot class loads a Machine Learning model and make inferences from input data.

public class SingleShot : IDisposable
Inheritance
Object
SingleShot
Implements
IDisposable

Constructors

View Source

SingleShot(String, NNFWType, HWType, Boolean)

Loads the neural network model and configures runtime environment without TensorsInfo

Declaration
public SingleShot(string modelAbsPath, NNFWType fwType = NNFWType.Any, HWType hwType = HWType.Any, bool isDynamicMode = false)
Parameters
Type Name Description
String modelAbsPath

Absolute path to the neural network model file.
NNFWType fwType

Types of Neural Network Framework (Default:NNFWType.Any)
HWType hwType

Types of hardware resources to be used for NNFWs (Default: HWType.Any)
Boolean isDynamicMode

Support Dynamic Mode (Default: false)
Exceptions
Type Condition
ArgumentException

Thrown when the method failed due to an invalid parameter.
NotSupportedException

Thrown when the feature is not supported.
API Level: 8
Feature: http://tizen.org/feature/machine_learning.inference
View Source

SingleShot(String, TensorsInfo, TensorsInfo)

Loads the neural network model and configures runtime environment

Declaration
public SingleShot(string modelAbsPath, TensorsInfo inTensorsInfo, TensorsInfo outTensorsInfo)
Parameters
Type Name Description
String modelAbsPath

Absolute path to the neural network model file.
TensorsInfo inTensorsInfo

Input TensorsInfo object
TensorsInfo outTensorsInfo

Output TensorsInfo object for inference result
Exceptions
Type Condition
ArgumentException

Thrown when the method failed due to an invalid parameter.
NotSupportedException

Thrown when the feature is not supported.
API Level: 6
Feature: http://tizen.org/feature/machine_learning.inference
View Source

SingleShot(String, TensorsInfo, TensorsInfo, NNFWType, HWType, Boolean)

Loads the neural network model and configures runtime environment with Neural Network Framework and HW information

Declaration
public SingleShot(string modelAbsPath, TensorsInfo inTensorsInfo, TensorsInfo outTensorsInfo, NNFWType fwType, HWType hwType, bool isDynamicMode)
Parameters
Type Name Description
String modelAbsPath

Absolute path to the neural network model file.
TensorsInfo inTensorsInfo

Input TensorsInfo object
TensorsInfo outTensorsInfo

Output TensorsInfo object for inference result
NNFWType fwType

Types of Neural Network Framework
HWType hwType

Types of hardware resources to be used for NNFWs
Boolean isDynamicMode

Support Dynamic Mode
Exceptions
Type Condition
ArgumentException

Thrown when the method failed due to an invalid parameter.
NotSupportedException

Thrown when the feature is not supported.
API Level: 8
Feature: http://tizen.org/feature/machine_learning.inference

Properties

View Source

Input

The information (tensor dimension, type, name and so on) of required input data for the given model.

Declaration
public TensorsInfo Input { get; set; }
Property Value
Type Description
TensorsInfo
Exceptions
Type Condition
NotSupportedException

Thrown when the feature is not supported.
ArgumentException

Thrown when the method failed due to an invalid parameter.
API Level: 8
View Source

Output

The information (tensor dimension, type, name and so on) of output data for the given model.

Declaration
public TensorsInfo Output { get; }
Property Value
Type Description
TensorsInfo
Exceptions
Type Condition
NotSupportedException

Thrown when the feature is not supported.
API Level: 8

Methods

View Source

Dispose()

Releases any unmanaged resources used by this object.

Declaration
public void Dispose()
API Level: 6
View Source

Dispose(Boolean)

Releases any unmanaged resources used by this object. Can also dispose any other disposable objects.

Declaration
protected virtual void Dispose(bool disposing)
Parameters
Type Name Description
Boolean disposing

If true, disposes any disposable objects. If false, does not dispose disposable objects.
View Source

Finalize()

Destructor of the Single instance.

Declaration
protected void Finalize()
API Level: 6
View Source

GetValue(String)

Gets the property value for the given model.

Declaration
public string GetValue(string name)
Parameters
Type Name Description
String name

The property name
Returns
Type Description
String

The property value
Exceptions
Type Condition
NotSupportedException

Thrown when the feature is not supported, or given property is not available.
ArgumentException

Thrown when the method failed due to an invalid parameter.
API Level: 8
Feature: http://tizen.org/feature/machine_learning.inference
View Source

Invoke(TensorsData)

Invokes the model with the given input data.

Declaration
public TensorsData Invoke(TensorsData inTensorsData)
Parameters
Type Name Description
TensorsData inTensorsData

The input data to be inferred.
Returns
Type Description
TensorsData

TensorsData instance which contains the inferred result.
Exceptions
Type Condition
ArgumentException

Thrown when the method failed due to an invalid parameter.
TimeoutException

Thrown when failed to get the result from sink element.
NotSupportedException

Thrown when the feature is not supported.
API Level: 6
Feature: http://tizen.org/feature/machine_learning.inference
View Source

SetTimeout(Int32)

Sets the maximum amount of time to wait for an output, in milliseconds.

Declaration
public void SetTimeout(int ms)
Parameters
Type Name Description
Int32 ms

The time to wait for an output (milliseconds)
Exceptions
Type Condition
NotSupportedException

Thrown when the feature is not supported.
ArgumentException

Thrown when the method failed due to an invalid parameter.
API Level: 8
Feature: http://tizen.org/feature/machine_learning.inference
View Source

SetValue(String, String)

Sets the property value for the given model.

A model/framework may support changing the model information, such as tensor dimension and data layout, after opening the model.

If tries to change unavailable property or the model does not allow changing the information, this will raise an exception.

For the details about the properties, see 'tensor_filter' plugin definition in NNStreamer.

Declaration
public void SetValue(string name, string value)
Parameters
Type Name Description
String name

The property name
String value

The property value
Exceptions
Type Condition
NotSupportedException

Thrown when the feature is not supported, or given property is not available.
ArgumentException

Thrown when the method failed due to an invalid parameter.
API Level: 8
Feature: http://tizen.org/feature/machine_learning.inference

Implements