Package | Description |
---|---|
org.nd4j.linalg.learning | |
org.nd4j.linalg.learning.config |
Modifier and Type | Interface and Description |
---|---|
interface |
GradientUpdater<T extends IUpdater>
Gradient modifications: Calculates an update and tracks related information for gradient changes over time
for handling updates.
|
Modifier and Type | Class and Description |
---|---|
class |
AdaDelta
http://www.matthewzeiler.com/pubs/googleTR2012/googleTR2012.pdf
https://arxiv.org/pdf/1212.5701v1.pdf
|
class |
AdaGrad
Vectorized Learning Rate used per Connection Weight
Adapted from: http://xcorr.net/2014/01/23/adagrad-eliminating-learning-rates-in-stochastic-gradient-descent/
See also http://cs231n.github.io/neural-networks-3/#ada
|
class |
Adam
The Adam updater.
|
class |
AdaMax
The AdaMax updater, a variant of Adam.
|
class |
Nadam
Setup and DynamicCustomOpsBuilder for Nadam updater.
|
class |
Nesterovs
Nesterov's momentum.
|
class |
NoOp
NoOp updater: gradient updater that makes no changes to the gradient
|
class |
RmsProp
RMS Prop updates:
|
class |
Sgd
SGD updater applies a learning rate only
|
Modifier and Type | Method and Description |
---|---|
IUpdater |
IUpdater.clone()
Clone the updater
|
IUpdater |
AdaMax.clone() |
Copyright © 2018. All rights reserved.