Class

org.platanios.tensorflow.api.ops.variables

VariableScope

Related Doc: package variables

Permalink

case class VariableScope extends Product with Serializable

Variable scope that carries default settings to provide to getVariable.

A variable scope allows to create new variables and to share already created ones while providing checks to not create or share by accident.

Many of the arguments we need for getVariable in a variable store are most easily handled with a context. VariableScope objects are used for the defaults.

Linear Supertypes

Serializable, Serializable, Product, Equals, AnyRef, Any

Ordering

Alphabetic
By Inheritance

Inherited

VariableScope
Serializable
Serializable
Product
Equals
AnyRef
Any

Hide All
Show All

Visibility

Public
All

Value Members

final def !=(arg0: Any): Boolean

Definition Classes
AnyRef → Any
final def ##(): Int

Definition Classes
AnyRef → Any
final def ==(arg0: Any): Boolean

Definition Classes
AnyRef → Any
final def asInstanceOf[T0]: T0

Definition Classes
Any
val cachingDevice: (OpSpecification) ⇒ String

Default caching device passed to getVariable.
def clone(): AnyRef

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( ... )
val dataType: types.DataType

Default data type passed to getVariable.
final def eq(arg0: AnyRef): Boolean

Definition Classes
AnyRef
def finalize(): Unit

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( classOf[java.lang.Throwable] )
final def getClass(): Class[_]

Definition Classes
AnyRef → Any
def getPartitionedVariable(store: VariableStore, name: String, dataType: types.DataType = this.dataType, shape: core.Shape = null, initializer: Initializer = this.initializer, regularizer: Regularizer = this.regularizer, partitioner: Partitioner = this.partitioner, trainable: Boolean = true, reuse: Reuse = this.reuse, collections: Set[Key[Variable]] = Set.empty, cachingDevice: (OpSpecification) ⇒ String = this.cachingDevice): PartitionedVariable

Gets an existing partitioned variable with the specified name or creates a new one.
Gets an existing partitioned variable with the specified name or creates a new one.
store
Variable store currently being used to store variables.
name
Variable name.
dataType
Variable data type.
shape
Variable shape.
initializer
Variable initializer. If initializer is null (the default), the default initializer passed in the constructor is used. If that one is null too, then we use a new glorotUniformInitializer. The initializer will be called for each part of the partitioned variable separately.
regularizer
Variable regularizer.
partitioner
Function that accepts a fully defined Shape and returns a sequence of integers (i.e., the partitions). These integers describe how to partition the given variable, along the each dimension. That is, partitions(1) = 3 means that we split the variable into 3 parts along dimension 1. Currently, partitioning along only a single axis is supported.
trainable
If true, the default, the variable is added to the graph collection Graph.Keys.TRAINABLE_VARIABLES. This collection is used as the default set of variables to use by the optimizers.
reuse
Reuse value indicating whether to re-use an existing variable with the same name, create a new variable, or do either.
collections
Set of graph collections keys. The variable is added to these collections. Defaults to Set(Graph.Keys.GLOBAL_VARIABLES).
cachingDevice
Device specification describing where the variable should be cached for reading. Defaults to the variable's device. Typical use is to cache on the device where the ops using the variable reside, to deduplicate copying through Switch and other conditional statements.
returns
Requested variable.

Annotations
@throws( ... ) @throws( ... ) @throws( ... )
Exceptions thrown
IllegalArgumentException If any of the provided arguments are not compatible with each other, or with the variables stored in this variable store.
InvalidDataTypeException If the provided data type does not match the data type of the corresponding variable stored in this variable store (if there exists one).
ShapeMismatchException If the provided shape does not match the shape of the corresponding variable stored in this variable store (if there exists one).
def getVariable(store: VariableStore, name: String, dataType: types.DataType = this.dataType, shape: core.Shape = null, initializer: Initializer = this.initializer, regularizer: Regularizer = this.regularizer, trainable: Boolean = true, reuse: Reuse = this.reuse, collections: Set[Key[Variable]] = Set.empty, cachingDevice: (OpSpecification) ⇒ String = this.cachingDevice): Variable

Gets an existing variable with the specified name or creates a new one.
Gets an existing variable with the specified name or creates a new one.
store
Variable store currently being used to store variables.
name
Variable name.
dataType
Variable data type.
shape
Variable shape.
initializer
Variable initializer. If initializer is null (the default), the default initializer passed in the constructor is used. If that one is null too, then we use a new glorotUniformInitializer. The initializer will be called for each part of the partitioned variable separately.
regularizer
Variable regularizer.
trainable
If true, the default, the variable is added to the graph collection Graph.Keys.TRAINABLE_VARIABLES. This collection is used as the default set of variables to use by the optimizers.
reuse
Reuse value indicating whether to re-use an existing variable with the same name, create a new variable, or do either.
- If reuse is null (the default), both new and existing variables are returned.
collections
Set of graph collections keys. The variable is added to these collections. Defaults to Set(Graph.Keys.GLOBAL_VARIABLES).
cachingDevice
Device specification describing where the variable should be cached for reading. Defaults to the variable's device. Typical use is to cache on the device where the ops using the variable reside, to deduplicate copying through Switch and other conditional statements.
returns
Requested variable.
Annotations
@throws( ... ) @throws( ... ) @throws( ... )
Exceptions thrown
IllegalArgumentException If any of the provided arguments are not compatible with each other, or with the variables stored in this variable store.
InvalidDataTypeException If the provided data type does not match the data type of the corresponding variable stored in this variable store (if there exists one).
ShapeMismatchException If the provided shape does not match the shape of the corresponding variable stored in this variable store (if there exists one).
val initializer: Initializer

Default initializer passed to getVariable.
final def isInstanceOf[T0]: Boolean

Definition Classes
Any
val name: String

Name of the variable scope, used as a prefix in getVariable.
val nameScope: String

Default name scope passed to getVariable.
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
val partitioner: Partitioner

Default partitioner passed to getVariable.
val regularizer: Regularizer

Default regularizer passed to getVariable.
val reuse: Reuse

Reuse value indicating whether to re-use an existing variable with the same name, create a new variable, or do either.
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
val underlyingGetter: VariableGetter

Default underlying variable getter passed to getVariable.
final def wait(): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )
final def wait(arg0: Long, arg1: Int): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )
final def wait(arg0: Long): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )

Inherited from Serializable

Inherited from Serializable

Inherited from Product

Inherited from Equals

Inherited from AnyRef

Inherited from Any

Ungrouped