Class Backend

Keras is a model-level library, providing high-level building blocks for developing deep learning models. It does not handle low-level operations such as tensor products, convolutions and so on itself. Instead, it relies on a specialized, well optimized tensor manipulation library to do so, serving as the "backend engine" of Keras. Rather than picking one single tensor library and making the implementation of Keras tied to that library, Keras handles the problem in a modular way, and several different backend engines can be plugged seamlessly into Keras.

Inheritance

System.Object

Keras

Base

Backend

Implements

System.IDisposable

Inherited Members

Base.Parameters

Base.None

Base.Init()

Base.ToPython()

Base.InvokeStaticMethod(Object, String, Dictionary<String, Object>)

Base.InvokeMethod(String, Dictionary<String, Object>)

Base.Item[String]

Keras.Instance

Keras.keras

Keras.keras2onnx

Keras.tfjs

Keras.Dispose()

Keras.ToTuple(Array)

Keras.ToList(Array)

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: Keras

Assembly: Keras.dll

Syntax

public class Backend : Base, IDisposable

Methods

| Improve this Doc View Source

CastToFloatX(NDarray)

Cast a Numpy array to the default Keras float type.

Declaration

public NDarray CastToFloatX(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	The x, Numpy array..

Returns

Type	Description
Numpy.NDarray	The same Numpy array, cast to its new type.

| Improve this Doc View Source

ClearSession()

Destroys the current TF graph and creates a new one.Useful to avoid clutter from old models / layers.

Declaration

public static void ClearSession()

| Improve this Doc View Source

Epsilon()

Returns the value of the fuzz factor used in numeric expressions.

Declaration

public static float Epsilon()

Returns

Type	Description
System.Single

| Improve this Doc View Source

FloatX()

Returns the default float type, as a string. (e.g. 'float16', 'float32', 'float64').

Declaration

public string FloatX()

Returns

Type	Description
System.String	String, the current default float type.

| Improve this Doc View Source

GetBackend()

Publicly accessible method for determining the current backend.

Declaration

public static string GetBackend()

Returns

Type	Description
System.String	String, the name of the backend Keras is currently using.

| Improve this Doc View Source

GetUid(String)

Get the uid for the default graph.

Declaration

public static string GetUid(string prefix = "")

Parameters

Type	Name	Description
System.String	prefix	An optional prefix of the graph.

Returns

Type	Description
System.String	A unique identifier for the graph.

| Improve this Doc View Source

ImageDataFormat()

Returns the default image data format convention.

Declaration

public static string ImageDataFormat()

Returns

Type	Description
System.String	A string, either 'channels_first' or 'channels_last'

| Improve this Doc View Source

IsSparse(NDarray)

Determines whether the specified tensor is sparse.

Declaration

public static bool IsSparse(NDarray tensor)

Parameters

Type	Name	Description
Numpy.NDarray	tensor	The tensor.

Returns

Type	Description
System.Boolean	`true` if the specified tensor is sparse; otherwise, `false`.

| Improve this Doc View Source

ResetUids()

Resets graph identifiers.

Declaration

public static void ResetUids()

| Improve this Doc View Source

SetEpsilon(Single)

Sets the value of the fuzz factor used in numeric expressions.

Declaration

public void SetEpsilon(float e)

Parameters

Type	Name	Description
System.Single	e	float. New value of epsilon.

| Improve this Doc View Source

SetFloatX(String)

Sets the default float type.

Declaration

public void SetFloatX(string floatx)

Parameters

Type	Name	Description
System.String	floatx	String, 'float16', 'float32', or 'float64'.

| Improve this Doc View Source

SetImageDataFormat(String)

Sets the image data format.

Declaration

public static void SetImageDataFormat(string data_format)

Parameters

Type	Name	Description
System.String	data_format	data_format: string. 'channels_first' or 'channels_last'.

| Improve this Doc View Source

ToDense(NDarray)

Converts a sparse tensor into a dense tensor and returns it.

Declaration

public static NDarray ToDense(NDarray tensor)

Parameters

Type	Name	Description
Numpy.NDarray	tensor	The tensor potentially sparse.

Returns

Type	Description
Numpy.NDarray	A dense tensor.

Implements

System.IDisposable