## Description

A function to predict testing data with built gradient descent model

## Usage

 1 prediction(model, dataTestInput) 

## Arguments

 model a matrix of coefficients used as a linear model to predict testing data input. This parameter exclusively produced by the gradient-descent-based learning function. dataTestInput a data.frame represented dataset with input variables only (m \times n-1), where m is the number of instances and n is the number of input variables only.

## Details

This function used to predict testing data with only input variable named dataTestInput. The model parameter is the coefficients that produced by gradient-descent-based learning function. The result of this function is a dataset that contains dataTestInput combined with prediction data as the last column of dataset.

## Value

a data.frame of testing data input variables and prediction variables.

GD, MBGD, SGD, SAGD, MGD, AGD, ADAGRAD, ADADELTA, RMSPROP, ADAM, SSGD, SVRG, SARAH, SARAHPlus
  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 ################################## ## Predict Testing Data Using GD Model ## load R Package data data(gradDescentRData) ## get z-factor Data dataSet <- gradDescentRData$CompressilbilityFactor ## do variance scaling to dataset featureScalingResult <- varianceScaling(dataSet) ## split dataset splitedDataSet <- splitData(featureScalingResult$scaledDataSet) ## built model using GD model <- GD(splitedDataSet$dataTrain) ## separate testing data with input only dataTestInput <- (splitedDataSet$dataTest)[,1:ncol(splitedDataSet$dataTest)-1] ## predict testing data using GD model prediction <- prediction(model,dataTestInput) ## show result() prediction ################################## ## Predict Testing Data Using SARAHPlus Model ## load R Package data data(gradDescentRData) ## get z-factor Data dataSet <- gradDescentRData$CompressilbilityFactor ## do variance scaling to dataset featureScalingResult <- varianceScaling(dataSet) ## split dataset splitedDataSet <- splitData(featureScalingResult$scaledDataSet) ## built model using SARAHPlus model <- SARAHPlus(splitedDataSet$dataTrain, alpha=0.1, maxIter=10, innerIter=10, gammaS=0.125, seed=NULL) ## separate testing data with input only dataTestInput <- (splitedDataSet$dataTest)[,1:ncol(splitedDataSet$dataTest)-1] ## predict testing data using GD model prediction <- prediction(model,dataTestInput) ## show result() prediction