Package deepnetts.net.train.opt

Optimization methods used by training algorithm.

See: Description

Interface Summary

Interface	Description
Optimizer	Optimization technique to tune network's weights parameters used by training algorithm.

Class Summary

Class	Description
AbstractOptimizer	Skeletal implementation of the Optimizer interface to minimize effort to implement specific optimizers.
AdaDeltaOptimizer	Implementation of ADADELTA Optimizer which is a modification od AdaGrad that uses only a limited window or previous gradients.
AdaGradOptimizer	Implementation of ADAGRAD Optimizer , which uses sum of squared previous gradients to adjust a global learning rate for each weight.
AdamOptimizer	Implementation of Adam optimizer which is a variation of RmsProp which includes momentum-like factor.
LearningRateDecay	https://www.coursera.org/learn/deep-neural-network/lecture/hjgIA/learning-rate-decay
MomentumOptimizer	Momentum optimization adds momentum parameter to basic Stochastic Gradient Descent, which can accelerate the process.
RmsPropOptimizer	A variation of AdaDelta optimizer.
SgdOptimizer	Basic Stochastic Gradient Descent optimization algorithm, which iteratively change weights towards value which gives minimum error.

Enum Summary

Enum	Description
OptimizerType	Supported types of optimization methods used by back-propagation training algorithm.

Package deepnetts.net.train.opt Description

Optimization methods used by training algorithm.