Class AbstractActivation

java.lang.Object

org.tensorflow.framework.activations.AbstractActivation

All Implemented Interfaces:

Activation

Direct Known Subclasses:

ELU, Exponential, GELU, HardSigmoid, Linear, ReLU, SELU, Sigmoid, Softmax, Softplus, Softsign, Swish, Tanh

public abstract class AbstractActivation
extends Object
implements Activation

Abstract base class for Activations

Field Summary

Fields

Modifier and Type	Field	Description
protected static final String	NAME_KEY
protected Ops	tf	The TensorFlow Ops

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	AbstractActivation()	Creates the abstract class for an AbstractActivation

Method Summary

Modifier and Type	Method	Description
protected void	checkClassName(Map<String,Object> config)	Verifies that the configuration is for the same Activation class.
protected void	checkConfigKeys(Set<String> keysToCheck, Set<String> allowedKeys)	Verifies that any key in keysToCheck is also in the allowedKeys set.
abstract Map<String,Object>	getConfig()	Gets a configuration map
protected Map<String,Object>	getDefaultConfig(String name)	Gets a configuration map, this default implementation returns a singleton Map, with NAME_KEY as the key, and the name parameter as its value;
abstract String	getName()	Get the name of the activation as known by the TensorFlow Engine
protected Ops	getTF()	Gets the TensorFlow Ops
protected void	setTF(Ops tf)	Sets the TensorFlow Ops

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface Activation

call

Modifier and Type	Method	Description
<T extends TNumber> Operand<T>	call(Ops tf, Operand<T> input)	Gets the calculation operation for the activation.

Field Details

NAME_KEY

protected static final String NAME_KEY

See Also:

• Constant Field Values

tf

protected Ops tf

The TensorFlow Ops

Constructor Details

AbstractActivation

protected AbstractActivation()

Creates the abstract class for an AbstractActivation

Method Details

getDefaultConfig

protected Map<String,Object> getDefaultConfig(String name)

Gets a configuration map, this default implementation returns a singleton Map, with NAME_KEY as the key, and the name parameter as its value;

Parameters:

name - the name of the Activation as known by TensorFlow engine.

Returns:

the configuration map

getConfig

public abstract Map<String,Object> getConfig()

Gets a configuration map

Returns:

the configuration map

getName

public abstract String getName()

Get the name of the activation as known by the TensorFlow Engine

Returns:

the name of the activation as known by the TensorFlow Engine

getTF

protected Ops getTF()

Gets the TensorFlow Ops

Returns:

the TensorFlow Ops

setTF

protected void setTF(Ops tf)

Sets the TensorFlow Ops

Parameters:

tf - the TensorFlow Ops

checkConfigKeys

protected void checkConfigKeys(Set<String> keysToCheck,
 Set<String> allowedKeys)

Verifies that any key in keysToCheck is also in the allowedKeys set.

Parameters:

keysToCheck - the set with keys to check

allowedKeys - the set to check against.

Throws:

IllegalArgumentException - if there is an entry in set1 that is not in set 2.

checkClassName

protected void checkClassName(Map<String,Object> config)

Verifies that the configuration is for the same Activation class.

Parameters:

config - the configuration

Throws:

IllegalArgumentException - if the value for the name key does not match the name for the Activation