Class WorkerFactory

Factory to create worker that executes specified model on a particular device (GPU, CPU, etc) using particular backend. See IWorker for usage of the worker itself.

Inheritance

object

WorkerFactory

BarracudaWorkerFactory

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Namespace: Unity.Barracuda

Assembly: Unity.Barracuda.dll

Syntax

public class WorkerFactory

Methods

CreateComputeWorker(Model, bool)

Create a worker using the precompiled GPU backend for the given model.

Declaration

public static IWorker CreateComputeWorker(Model model, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
bool	verbose

Returns

Type	Description
IWorker	Worker instance

CreatePixelShaderWorker(Model, bool)

Create a worker using the reference GPU backend for the given model.

Declaration

public static IWorker CreatePixelShaderWorker(Model model, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateReferenceCPUWorker(Model, bool)

Create a worker using the reference CPU backend for the given model.

Declaration

public static IWorker CreateReferenceCPUWorker(Model model, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateReferenceComputeWorker(Model, bool)

Create a worker using the reference GPU backend for the given model.

Declaration

public static IWorker CreateReferenceComputeWorker(Model model, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Model, bool)

Create a worker that will execute model using the best backend that is available for a given device type.

Declaration

public static IWorker CreateWorker(Model model, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
bool	verbose	will log scheduling of layers execution to the console

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Model, string[], string[], Device, bool)

Create a worker that will execute model using the best backend that is available for a given device type.

Declaration

public static IWorker CreateWorker(Model model, string[] additionalOutputs, string[] trimOutputs, WorkerFactory.Device device = Device.Auto, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
string[]	additionalOutputs	the additional outputs to track but not directly specified by the model
string[]	trimOutputs	by specifying this list of outputs, all other non-specified outputs will be discarded
WorkerFactory.Device	device	the device type to run worker on. For example WorkerFactory.Device.GPU specifies the fast GPU path
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Model, string[], Device, bool)

Create a worker that will execute model using the best backend that is available for a given device type.

Declaration

public static IWorker CreateWorker(Model model, string[] additionalOutputs, WorkerFactory.Device device = Device.Auto, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
string[]	additionalOutputs	the additional outputs to track but not directly specified by the model
WorkerFactory.Device	device	the device type to run worker on. For example WorkerFactory.Device.GPU specifies the fast GPU path
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Model, Device, bool)

Create a worker that will execute model using the best backend that is available for a given device type.

Declaration

public static IWorker CreateWorker(Model model, WorkerFactory.Device device, bool verbose = false)

Parameters

Type	Name	Description
Model	model	the associated model. See ModelLoader.cs
WorkerFactory.Device	device	the preferred device for execution. For example WorkerFactory.Device.GPU specifies the fast GPU path
bool	verbose	will log scheduling of layers execution to the console

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, bool)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, bool verbose)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
bool	verbose	will log scheduling of layers execution to the console

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, bool, Type, LogLevel)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, bool verbose, WorkerFactory.Type compareAgainstType, CompareOpsUtils.LogLevel differenceLogLevel = LogLevel.Warning)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
bool	verbose	will log scheduling of layers execution to the console
WorkerFactory.Type	compareAgainstType	if different than type model will be run on those two backend and result of every layer will be compared, checking for divergence. Great for debugging, but very slow because of the sync needed
CompareOpsUtils.LogLevel	differenceLogLevel	if compareAgainstType is used difference will be reported as error is this is true or warning otherwise

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, string[], bool)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, string[] additionalOutputs, bool verbose = false)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
string[]	additionalOutputs	the additional outputs to track but not directly specified by the model
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, string[], string[], bool)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, string[] additionalOutputs = null, string[] trimOutputs = null, bool verbose = false)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
string[]	additionalOutputs	the additional outputs to track but not directly specified by the model
string[]	trimOutputs	by specifying this list of outputs, all other non-specified outputs will be discarded
bool	verbose	will log scheduling of layers execution to the console (default == false)

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, string[], string[], bool, Type, LogLevel)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, string[] additionalOutputs, string[] trimOutputs, bool verbose, WorkerFactory.Type compareAgainstType, CompareOpsUtils.LogLevel differenceLogLevel = LogLevel.Warning)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
string[]	additionalOutputs	the additional outputs to track but not directly specified by the model
string[]	trimOutputs	by specifying this list of outputs, all other non-specified outputs will be discarded
bool	verbose	will log scheduling of layers execution to the console
WorkerFactory.Type	compareAgainstType	if different than type model will be run on those two backend and result of every layer will be compared, checking for divergence. Great for debugging, but very slow because of the sync needed
CompareOpsUtils.LogLevel	differenceLogLevel	if compareAgainstType is used difference will be reported as error is this is true or warning otherwise

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, string[], string[], WorkerConfiguration, IModelExecutionsReporter)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, string[] additionalOutputs, string[] trimOutputs, WorkerFactory.WorkerConfiguration workerConfiguration, IModelExecutionsReporter modelExecutionsReporter = null)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
string[]	additionalOutputs	the additional outputs to track but not directly specified by the model
string[]	trimOutputs	by specifying this list of outputs, all other non-specified outputs will be discarded
WorkerFactory.WorkerConfiguration	workerConfiguration	define configurations such as logging and comparison backend, see WorkerConfiguration API docs
IModelExecutionsReporter	modelExecutionsReporter	execution reporter to use to track models executions

Returns

Type	Description
IWorker	Worker instance

CreateWorker(Type, Model, WorkerConfiguration)

Create a worker with explicitly specified backend type to execute the given model.

Declaration

public static IWorker CreateWorker(WorkerFactory.Type type, Model model, WorkerFactory.WorkerConfiguration workerConfiguration)

Parameters

Type	Name	Description
WorkerFactory.Type	type	backend type to use. For example WorkerFactory.Type.Compute specifies the fast GPU path
Model	model	the associated model. See ModelLoader.cs
WorkerFactory.WorkerConfiguration	workerConfiguration	define configurations such as logging and comparison backend, see WorkerConfiguration API docs

Returns

Type	Description
IWorker	Worker instance

GetBestTypeForDevice(Device)

Returns the best backend type that can run on a device given the model.

Declaration

public static WorkerFactory.Type GetBestTypeForDevice(WorkerFactory.Device device)

Parameters

Type	Name	Description
WorkerFactory.Device	device	device

Returns

Type	Description
WorkerFactory.Type	Best worker type for specified device

IsType(Type, Device)

Check if a backend is of a given type. For example: IsType(Type.CSharpRef, Device.GPU) == true

Declaration

public static bool IsType(WorkerFactory.Type type, WorkerFactory.Device device)

Parameters

Type	Name	Description
WorkerFactory.Type	type	type to check against
WorkerFactory.Device	device	device to check against

Returns

Type	Description
bool	true if backend is of specified type

Exceptions

Type	Condition
ArgumentException	thrown if type is Type.Auto

ValidateType(Type)

Validate if a backend of type is supported, otherwise return a fallback type.

Declaration

public static WorkerFactory.Type ValidateType(WorkerFactory.Type type)

Parameters

Type	Name	Description
WorkerFactory.Type	type	type

Returns

Type	Description
WorkerFactory.Type	returns type if valid, otherwise returns fallback type