backpropagate_delta_bn

darch

trainData

targetData

error function to minimize during training. Right now mean squared
erros and cross entropy errors are supported.

errorFunc

traing with batch normalization on or off

with_BN


function that calculates the delta function of a darch object with batch
normalization


An implementation of deep neural network with rectifier linear units trained
with stochastic gradient descent method and batch normalization. A combination of these methods have
achieved state-of-the-art performance in ImageNet classification by overcoming the gradient saturation
problem experienced by many deep architecture neural network models in the past. In addition,
batch normalization and dropout are implemented as a means of regularization. The deeplearning package is
inspired by the darch package and uses its class DArch.

backpropagate_delta_bn: Calculates the delta functions using backpropagation

Description

Usage

Arguments

References

See Also