Last chance! 50% off unlimited learning
Sale ends in
A function to build prediction model using RMSPROP method.
RMSPROP(dataTrain, alpha = 0.1, maxIter = 10, momentum = 0.9,
seed = NULL)
a data.frame that representing training data (
a float value representing learning rate. Default value is 0.1
the maximal number of iterations.
a float value represent momentum give a constant speed to learning process.
a integer value for static random. Default value is NULL, which means the function will not do static random.
a vector matrix of theta (coefficient) for linear model.
This function based on SGD
with an optimization to create
an adaptive learning rate by RMS cost and hessian approximation correction approach.
In other word, this method combine the ADAGRAD
and ADADELTA
approaches.
M. D. Zeiler Adadelta: An Adaptive Learning Rate Method, arXiv: 1212.5701v1, pp. 1-6 (2012)
# NOT RUN {
##################################
## Learning and Build Model with RMSPROP
## load R Package data
data(gradDescentRData)
## get z-factor data
dataSet <- gradDescentRData$CompressilbilityFactor
## split dataset
splitedDataSet <- splitData(dataSet)
## build model with RMSPROP
RMSPROPmodel <- RMSPROP(splitedDataSet$dataTrain)
#show result
print(RMSPROPmodel)
# }
Run the code above in your browser using DataLab