SGD

Instance Constructors

new SGD(learningRate: Double = 1e-3, learningRateDecay: Double = 0.0, weightDecay: Double = 0.0, momentum: Double = 0.0, dampening: Double = Double.MaxValue, nesterov: Boolean = false, learningRateSchedule: LearningRateSchedule = Default(), learningRates: Tensor[T] = null, weightDecays: Tensor[T] = null)(implicit arg0: ClassTag[T], ev: TensorNumeric[T])

learningRate
learning rate
learningRateDecay
learning rate decay
weightDecay
weight decay
momentum
momentum
dampening
dampening for momentum
nesterov
enables Nesterov momentum
learningRates
1D tensor of individual learning rates
weightDecays
1D tensor of individual weight decays

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
def clearHistory(): Unit

Clear the history information in the OptimMethod state
Clear the history information in the OptimMethod state

Definition Classes
SGD → OptimMethod
def clone(): OptimMethod[T]

clone OptimMethod
clone OptimMethod

Definition Classes
OptimMethod → AnyRef
var dampening: Double

dampening for momentum
final def eq(arg0: AnyRef): Boolean

Definition Classes
AnyRef
def equals(arg0: Any): Boolean

Definition Classes
AnyRef → Any
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 getHyperParameter(config: Table): String

return an string of current hyperParameter.
return an string of current hyperParameter.
config
a table contains the hyper parameter.

Definition Classes
SGD → OptimMethod
def getHyperParameter(): String

return an string of current hyperParameter.
return an string of current hyperParameter.

Definition Classes
SGD → OptimMethod
def getLearningRate(): Double

get learning rate
get learning rate

Definition Classes
SGD → OptimMethod
def hashCode(): Int

Definition Classes
AnyRef → Any
final def isInstanceOf[T0]: Boolean

Definition Classes
Any
var learningRate: Double

learning rate
var learningRateDecay: Double

learning rate decay
var learningRateSchedule: LearningRateSchedule
var learningRates: Tensor[T]

1D tensor of individual learning rates
def loadFromTable(config: Table): SGD.this.type

load optimMethod parameters from Table
load optimMethod parameters from Table

Definition Classes
SGD → OptimMethod
var momentum: Double

momentum
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
var nesterov: Boolean

enables Nesterov momentum
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
def optimize(feval: (Tensor[T]) ⇒ (T, Tensor[T]), x: Tensor[T]): (Tensor[T], Array[T])

feval
a function that takes a single input (X), the point of a evaluation, and returns f(X) and df/dX
x
the initial point
returns
the new x 1D tensor and the function list, evaluated before the update

Definition Classes
SGD → OptimMethod
def save(path: String, overWrite: Boolean = false): SGD.this.type

save OptimMethod
save OptimMethod
path
path
overWrite
whether to overwrite

Definition Classes
OptimMethod
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
def toString(): String

Definition Classes
AnyRef → Any
def updateHyperParameter(config: Table, state: Table): Unit

Update hyper parameter.
Update hyper parameter. We have updated hyper parameter in method optimize(). But in DistriOptimizer, the method optimize() is only called on the executor side, the driver's hyper parameter is unchanged. So this method is using to update hyper parameter on the driver side.
config
config table.
state
state Table.
returns
A string.

Definition Classes
SGD → OptimMethod
def updateHyperParameter(): Unit

Update hyper parameter.
Update hyper parameter. We have updated hyper parameter in method optimize(). But in DistriOptimizer, the method optimize() is only called on the executor side, the driver's hyper parameter is unchanged. So this method is using to update hyper parameter on the driver side.
returns
A string.

Definition Classes
SGD → OptimMethod
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( ... )
var weightDecay: Double

weight decay
var weightDecays: Tensor[T]

1D tensor of individual weight decays

Deprecated Value Members

def clearHistory(state: Table): Table

Clear the history information in the state
Clear the history information in the state

Definition Classes
OptimMethod
Annotations
@deprecated
Deprecated
(Since version 0.2.0) Please use clearHistory() instead
def optimize(feval: (Tensor[T]) ⇒ (T, Tensor[T]), parameter: Tensor[T], config: Table, state: Table = null): (Tensor[T], Array[T])

Optimize the model parameter
Optimize the model parameter
feval
a function that takes a single input (X), the point of a evaluation, and returns f(X) and df/dX
parameter
the initial point
config
a table with configuration parameters for the optimizer
state
a table describing the state of the optimizer; after each call the state is modified
returns
the new x vector and the function list, evaluated before the update

Definition Classes
OptimMethod
Annotations
@deprecated
Deprecated
(Since version 0.2.0) Please initialize OptimMethod with parameters when creating it instead of importing table

Related Docs: object SGD | package optim

class SGD[T] extends OptimMethod[T]

Instance Constructors

Value Members

final def !=(arg0: Any): Boolean

final def ##(): Int

final def ==(arg0: Any): Boolean

final def asInstanceOf[T0]: T0

def clearHistory(): Unit

def clone(): OptimMethod[T]

var dampening: Double

final def eq(arg0: AnyRef): Boolean

def equals(arg0: Any): Boolean

def finalize(): Unit

final def getClass(): Class[_]

def getHyperParameter(config: Table): String

def getHyperParameter(): String

def getLearningRate(): Double

def hashCode(): Int

final def isInstanceOf[T0]: Boolean

var learningRate: Double

var learningRateDecay: Double

var learningRateSchedule: LearningRateSchedule

var learningRates: Tensor[T]

def loadFromTable(config: Table): SGD.this.type

var momentum: Double

final def ne(arg0: AnyRef): Boolean

var nesterov: Boolean

final def notify(): Unit

final def notifyAll(): Unit

def optimize(feval: (Tensor[T]) ⇒ (T, Tensor[T]), x: Tensor[T]): (Tensor[T], Array[T])

def save(path: String, overWrite: Boolean = false): SGD.this.type

final def synchronized[T0](arg0: ⇒ T0): T0

def toString(): String

def updateHyperParameter(config: Table, state: Table): Unit

def updateHyperParameter(): Unit

final def wait(): Unit

final def wait(arg0: Long, arg1: Int): Unit

final def wait(arg0: Long): Unit

var weightDecay: Double

var weightDecays: Tensor[T]

Deprecated Value Members

def clearHistory(state: Table): Table

def optimize(feval: (Tensor[T]) ⇒ (T, Tensor[T]), parameter: Tensor[T], config: Table, state: Table = null): (Tensor[T], Array[T])

Inherited from OptimMethod[T]

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped