deepforest {deepdive}

Build or train bagged deeptree or deepnet of multiple architecture

Description

Build or train bagged deeptree or deepnet of multiple architecture.Based on error choice either select best model or average multiple model with random variable cut,data cut and architechture

Usage

deepforest(
  x,
  y,
  networkCount = 3,
  layerChoice = c(2:3),
  unitsChoice = c(4:10),
  cutVarSizePercent = 0.6,
  cutDataSizePercent = 0.6,
  activation = c("sigmoid", "sigmoid"),
  reluLeak = 0,
  modelType = "regress",
  iterations = 500,
  eta = 10^-2,
  seed = 2,
  gradientClip = 0.8,
  regularisePar = 0,
  optimiser = "adam",
  parMomentum = 0.9,
  inputSizeImpact = 1,
  parRmsPropZeroAdjust = 10^-8,
  parRmsProp = 0.9999,
  treeLeaves = NA,
  treeMinSplitPercent = 0.3,
  treeMinSplitCount = 100,
  treeCp = 0.01,
  errorCover = 0.2,
  treeAugment = TRUE,
  printItrSize = 100,
  showProgress = TRUE,
  stopError = 0.01,
  miniBatchSize = NA,
  useBatchProgress = TRUE
)

Arguments

x

a data frame with input variables

y

a data frame with ouptut variable

networkCount

Integer, Number of deepnet or deeptree to build

layerChoice

vector, different layer choices

unitsChoice

vector , number of units choice

cutVarSizePercent

ratio, percentage of variable to for each network

cutDataSizePercent

ratio, percentage of data to for each network

activation

choose from "sigmoid","relu","sin","cos","none".Activations will be randomly chosen from chosen. Default is relu and sin

reluLeak

numeric. Applicable when activation is "relu". Specify value between 0 any number close to zero below 1. Eg: 0.01,0.001 etc

modelType

one of "regress","binary","multiClass". "regress" for regression will create a linear single unit output layer. "binary" will create a single unit sigmoid activated layer. "multiClass" will create layer with units corresponding to number of output classes with softmax activation.

iterations

integer. This indicates number of iteratios or epochs in backpropagtion .The default value is 500.

eta

numeric.Hyperparameter,sets the Learning rate for backpropagation. Eta determines the convergence ability and speed of convergence.

seed

numeric. Set seed with this parameter. Incase of sin activation sometimes changing seed can yeild better results. Default is 2

gradientClip

numeric. Hyperparameter numeric value which limits gradient size for weight update operation in backpropagation. Default is 0.8 . It can take any postive value.

regularisePar

numeric. L2 Regularisation Parameter .

optimiser

one of "gradientDescent","momentum","rmsProp","adam". Default value "adam"

parMomentum

numeric. Applicable for optimiser "mometum" and "adam"

inputSizeImpact

numeric. Adjusts the gradient size by factor of percentage of rows in input. For very small data set setting this to 0 could yeild faster result. Default is 1.

parRmsPropZeroAdjust

numeric. Applicable for optimiser "rmsProp" and "adam"

parRmsProp

numeric.Applicable for optimiser "rmsProp" and "adam"

treeLeaves

vector.Optional , leaves numbers from externally trained tree model can be supplied here. If supplied then model will not build a explicit tree and just fit a neural network to mentioned leaves.

treeMinSplitPercent

numeric. This parameter controls depth of tree setting min split count for leaf subdivision as percentage of observations. Final minimum split will be chosen as max of count calculted with treeMinSplitPercent and treeMinSplitCount. Default 0.3. Range 0 to 1.

treeMinSplitCount

numeric. This parameter controls depth of tree setting min split count.Final minimum split will be chosen as max of count calculted with treeMinSplitPercent and treeMinSplitCount. Default 30

treeCp

complexity parameter. rpart.control

errorCover

Ratio. Deault is 0.2 i.e all models within 20 percent error of best model will be selected.

treeAugment

logical. If True fits deeptree and if False fits deepnet. Default is T

printItrSize

numeric. Number of iterations after which progress message should be shown. Default value 100 and for iterations below 100 atleast 5 messages will be seen

showProgress

logical. True will show progress and F will not show progress

stopError

Numeric. Rmse at which iterations can be stopped. Default is 0.01, can be set as NA in case all iterations needs to run.

miniBatchSize

integer. Set the mini batch size for mini batch gradient

useBatchProgress

logical. Applicable for miniBatch , setting T will use show rmse in Batch and F will show error on full dataset. For large dataset set T

Value

returns model object which can be passed into predict.deepforest

Examples

require(deepdive)

x<-data.frame(x1=runif(10),x2=runif(10))
y<-data.frame(y=10*x$x1+20*x$x2+20)

mdeepf<-deepforest(x,y,
                  networkCount=2,
                  layerChoice=c(2:3),
                  unitsChoice=c(4:10),
                  cutVarSizePercent=0.6,
                  cutDataSizePercent=0.6,
                  activation = c('relu',"sin"),
                  reluLeak=0.01,
                  modelType ='regress',
                  iterations = 10,
                  eta = 10 ^-2,
                  seed=2,
                  gradientClip=0.8,
                  regularisePar=0,
                  optimiser="adam",
                  parMomentum=0.9,
                  inputSizeImpact=1,
                  parRmsPropZeroAdjust=10^-8,
                  parRmsProp=0.9999,
                  treeLeaves=NA,
                  treeMinSplitPercent=0.3,
                  treeMinSplitCount=100,
                  treeCp=0.01 ,
                  errorCover=0.2,
                  treeAugment=TRUE,
                  printItrSize=100,
                  showProgress=TRUE,
                  stopError=0.01,
                  miniBatchSize=64,
                  useBatchProgress=TRUE)