library(FactoMineR)
library(factoextra)
library(MASS)
library(fields)
library(plotly)
library(mgcv)
library(MCMCpack)
library(glmulti)
library(plyr)
library(lattice)
library(ggplot2)
library(reshape2)
library(shiny)
library(googleVis)
library(multcomp)
library(sjPlot)
library(scales)
library(doBy)
library(gridExtra)
library(agricolae)
library(devtools)
library(magrittr)
library(ggbiplot)
library(ggdendro)
library(dendextend)
library(grid)
#devtools::install_github("vqv/ggbiplot")
#install.packages("ggdendro")
#install.packages("pastecs")
library(pastecs)
library(shinydashboard);
source('global.R',local = TRUE)
shinyServer(function(input,output,session)
{
#### download data
rawInputData0<- reactive({
rawData = input$rawInputFile
headerTag = input$headerUI;
sepTag = input$sepUI;
quoteTag = input$quoteUI;
decTag = input$decUI;
if(!is.null(rawData)) {
data = read.csv(rawData$datapath,header=headerTag,dec=decTag ,sep=sepTag,quote=quoteTag);
} else {
return(NULL);
}
});
output$product <- renderUI({
data = rawInputData0();
if (is.null(data)) return(NULL)
items=names(data)
names(items)=items
selectInput("product","Select product:",items);
})
output$assessor <- renderUI({
data = rawInputData0();
if (is.null(data)) return(NULL)
items=names(data)
names(items)=items
selectInput("assessor","Select assessor:",items,multiple = T)
})
output$replication <- renderUI({
data = rawInputData0();
if (is.null(data)) return(NULL)
items=names(data)
names(items)=items
selectInput("replication","Select replication:",items,multiple = T)
})
col=reactive({
dat=rawInputData0()
s=0
n=which(colnames(dat) %in% c(paste(input$product),paste(input$replication),paste(input$assessor)))
for(i in 1:ncol(dat)){
if (i %in% n)
s[i]="factor"
else s[i]=class(dat[,i])
}
return(s)
})
rawInputData<- reactive({
rawData = input$rawInputFile
headerTag = input$headerUI;
sepTag = input$sepUI;
quoteTag = input$quoteUI;
decTag = input$decUI;
if(!is.null(rawData)) {
data = read.csv(rawData$datapath,header=headerTag,sep=sepTag,quote=quoteTag,dec=decTag,colClasses = col());
} else {
return(NULL);
}
});
consInputData = reactive({
rd = input$consInputFile
ht = input$headUI;
st = input$sepaUI;
qt = input$quotUI;
decc = input$deccUI;
if(!is.null(rd)) {
data = read.csv(rd$datapath,header=ht,sep=st,dec=decc,quote=qt,row.names = 1);
} else {
return(NULL);
}
});
output$stat_sum <- renderDataTable({
data = rawInputData()
df.data = stat.desc(data)
format(df.data,digits=2)
})
output$pre.data <- renderDataTable({
data = rawInputData()
df.data = data.frame(data)
df.data
})
output$str <- renderPrint({
dat <- rawInputData()
str(dat)
})
moy=reactive({
data=rawInputData()
base=summaryBy(as.formula(paste(". ~ ",paste(input$product,collapse="+"))),data=data,FUN=c(mean),id=paste(input$origine),na.rm=T,keep.names=T)
base=data.frame(base,row.names =1)
base
})
moy1=reactive({
data=rawInputData()
base=summaryBy(as.formula(paste(". ~ ",paste(input$product,collapse="+"))),data=data,FUN=c(mean),na.rm=T,keep.names=T)
base
})
moy2=reactive({
data=rawInputData()
if(!is.null(data)){
base=summaryBy(as.formula(paste(". ~ ",paste(input$product,collapse="+"))),data=data,FUN=c(mean),na.rm=T,keep.names=T)
base=data.frame(base,row.names =1)
if(!is.null(physicalData())){
base=cbind.data.frame(base,physicalData())
base}
else{base}}
else{base=data.frame(physicalData())
base}
})
mo=reactive({
data=rawInputData()
if(!is.null(data)){
base=summaryBy(as.formula(paste(". ~ ",paste(input$product,collapse="+"))),data=data,FUN=c(mean),na.rm=T,keep.names=T)
base=data.frame(base,row.names =1)
if(!is.null(physicalData())){
base=cbind.data.frame(base,physicalData())
av=format(base,digits = 3,decimal.mark = ".")}
else{av=format(base,digits = 3,decimal.mark = ".")}}
else{base=data.frame(physicalData())
av=format(base,digits = 3,decimal.mark = ".")}
})
moy3=reactive({
data=rawInputData()
nums <- sapply(data, is.factor)
items=names(nums[nums])
df=summaryBy(as.formula(paste(". ~ ",paste(input$product,collapse="+"))),data=data,FUN=c(mean),na.rm=T,keep.names=T,id= items )
df=data.frame(df)
})
output$av<- renderDataTable({
mo()
})
output$downloadav <- downloadHandler(
filename = paste0("average_", Sys.Date(),".csv"),
content = function(file) {
write.csv(mo(),file,row.names=F)
}
)
output$av.table<- renderDataTable({
moy2()
})
moy4=reactive({
if(!is.null(input$assessor)){
data=rawInputData()
base=summaryBy(as.formula(paste(". ~ ",paste(input$product,"+",input$assessor))),data=data,FUN=c(mean),na.rm=T,keep.names=T)
base=base[,-3]}
else if (is.null(input$assessor)) return(NULL)
})
output$av.table2<- renderDataTable({
moy4()
})
output$downloadavt <- downloadHandler(
filename = paste0("averagejuge_", Sys.Date(),".csv"),
content = function(file) {
write.csv(moy4(),file,row.names=T)
}
)
cercle <- reactive({
base=moy2()
res <-PCA(base,ncp=2)
fviz_pca_var(res)+
scale_color_gradient2(low="grey", mid="blue",
high="red", midpoint=0.6, space = "Lab")+theme_minimal()
})
output$cercle <- renderPlot({
cercle()
})
output$downloadPlot1 <- downloadHandler(
filename = function() {
paste("cercle", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(cercle())
dev.off(which=dev.cur())}
)
scree <- reactive({
base=moy2()
res <-PCA(base,ncp=2)
fviz_eig(res, addlabels=TRUE, hjust = -0.3,
linecolor ="red") +
theme_minimal()
})
output$scree <- renderPlot({
scree()
})
output$downloadPlot2 <- downloadHandler(
filename = function() {
paste("scree", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(scree())
dev.off(which=dev.cur())}
)
stable1 <- reactive({
base=moy2()
res <-PCA(base,ncp=2)
fviz_screeplot(res, linecolor = "red",cex=0.7)
res$eig
})
output$stable <- renderTable({
stable1()
})
output$downloadstab <- downloadHandler(
filename = paste0("Eigen_", Sys.Date(),".csv"),
content = function(file) {
write.csv(stable1(),file,row.names=T)
}
)
ind <- reactive({
base=moy2()
res <-PCA(base,ncp=2)
fviz_pca_ind(res)
})
output$ind <- renderPlot({
ind()
})
output$downloadPlot3 <- downloadHandler(
filename = function() {
paste("ind", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(ind())
dev.off(which=dev.cur())}
)
bi <- reactive({
X=moy2()
res.pca = PCA(X,ncp = 3, graph = FALSE, scale.unit=1)
library(devtools)
library(ggbiplot)
pca <- prcomp(X, scale. = TRUE)
g <- ggbiplot(pca)
g <- g + scale_color_discrete(name = '')
g <- g + theme(legend.direction = 'horizontal',
legend.position = 'top')
print(g)
biplot=fviz_pca_biplot(res.pca,col.ind = "blue", col.var = "black", title="") + theme_bw()+ggtitle( "Biplot of variables and individuals on first PCA factor map")
biplot + theme(plot.title = element_text(size=16, hjust = 0.5,face="bold"))
})
output$bi <- renderPlot({
bi()
})
output$downloadPlot4 <- downloadHandler(
filename = function() {
paste("bi", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(bi())
dev.off(which=dev.cur())}
)
output$cons.data <- renderDataTable({
data = consInputData()
df.data = data.frame(data)
df.data
})
cerclec <- reactive({
base=consInputData()
res <-PCA(base,ncp=2)
fviz_pca_var(res)+theme_minimal()
})
output$cerclec <- renderPlot({
cerclec()
})
output$downloadPlot5 <- downloadHandler(
filename = function() {
paste("cerclec", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(cerclec())
dev.off(which=dev.cur())}
)
output$cerclec1 <- renderPlot({
base=consInputData()
res <-PCA(base,ncp=2)
fviz_pca_var(res)+
scale_color_gradient2(low="grey", mid="navy",
high="red", midpoint=96, space = "Lab")
})
screec <- reactive({
base=consInputData()
res <-PCA(base,ncp=2)
fviz_eig(res, addlabels=TRUE, hjust = -0.3,
linecolor ="red") +
theme_minimal()
})
output$screec <- renderPlot({
screec()
})
output$downloadPlot6 <- downloadHandler(
filename = function() {
paste("screec", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(screec())
dev.off(which=dev.cur())}
)
stablec1 <- reactive({
base=consInputData()
res <-PCA(base,ncp=2)
fviz_screeplot(res, linecolor = "red",cex=0.7)
res$eig
})
output$stablec <- renderTable({
stablec1()
})
output$downloadstab1 <- downloadHandler(
filename = paste0("Eigen_", Sys.Date(),".csv"),
content = function(file) {
write.csv(stablec1(),file,row.names=T)
}
)
indc <- reactive({
base=consInputData()
res <-PCA(base,ncp=2)
fviz_pca_ind(res)
})
output$indc <- renderPlot({
indc()
})
output$downloadPlot7 <- downloadHandler(
filename = function() {
paste("indc", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(indc())
dev.off(which=dev.cur())}
)
bic <- reactive({
X=consInputData()
res.pca = PCA(X,ncp = 3, graph = FALSE, scale.unit=1)
library(devtools)
library(ggbiplot)
pca <- prcomp(X, scale. = TRUE)
g <- ggbiplot(pca)
g <- g + scale_color_discrete(name = '')
g <- g + theme(legend.direction = 'horizontal',
legend.position = 'top')
print(g)
biplot=fviz_pca_biplot(res.pca,col.ind = "blue", col.var = "black", title="") + theme_bw()+ggtitle( "Biplot of variables and individuals on first PCA factor map")
biplot + theme(plot.title = element_text(size=16, hjust = 0.5,face="bold"))
})
output$bic <- renderPlot({
bic()
})
output$downloadPlot8 <- downloadHandler(
filename = function() {
paste("bic", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(bic())
dev.off(which=dev.cur())}
)
output$summary.data <- renderTable({
data = rawInputData()
summary(data)
})
output$variablee= renderUI({
df = rawInputData();
if (is.null(df)) return(NULL)
nums <- sapply(df, is.numeric)
items=names(nums[nums])
names(items)=items
selectInput("variablee","Select variable from:",items)
})
output$Min=renderValueBox({
df=rawInputData()
a=summary(df)
c=grep(input$variablee, colnames(df))
c=c[1]
valueBox(a[[1,c]],
"Minimum",
color = "red",
icon = icon("thumbs-o-down")
)
})
output$first=renderValueBox({
df=rawInputData()
a=summary(df)
c=grep(input$variablee, colnames(df))
c=c[1]
valueBox(a[[2,c]],
"1st Quartile",
color = "maroon",
icon = icon("list-ol")
)
})
output$Median=renderValueBox({
df=rawInputData()
a=summary(df)
c=grep(input$variablee, colnames(df))
c=c[1]
valueBox(a[[3,c]],
"Median",
color = "olive",
icon = icon("align-center")
)
})
output$Mean=renderValueBox({
df=rawInputData()
a=summary(df)
c=grep(input$variablee, colnames(df))
c=c[1]
valueBox(a[[4,c]],
"Mean",
color = "teal",
icon = icon("align-center")
)
})
output$third=renderValueBox({
df=rawInputData()
a=summary(df)
c=grep(input$variablee, colnames(df))
c=c[1]
valueBox(a[[5,c]],
"3rd Quartile",
color = "orange",
icon = icon("list-ol")
)
})
output$Max=renderValueBox({
df=rawInputData()
a=summary(df)
c=grep(input$variablee, colnames(df))
c=c[1]
valueBox(a[[6,c]],
"Maximum",
color = "purple",
icon = icon("thumbs-o-up")
)
})
output$var_qual <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("var_qual", "Select a variable :",items)
})
output$pp<- renderPrint({
df=rawInputData()
c=grep(input$var_qual, colnames(df))
data=df[,c]
a=sjt.frq(data)
return(HTML(a$knitr))
})
output$var_qual<- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("var_qual", "Select a variable :",items)
})
output$pp_qual<- renderPrint({
df=rawInputData()
c=grep(input$var_qual, colnames(df))
data=df[,c]
a=sjt.frq(data)
return(HTML(a$knitr))
})
output$summary.data1 <- renderTable({
data = consInputData()
summary(data)
})
output$variablee1= renderUI({
df = consInputData();
if (is.null(df)) return(NULL)
nums <- sapply(df, is.numeric)
items=names(nums[nums])
names(items)=items
selectInput("variablee1","Select variable from:",items)
})
output$Min1=renderValueBox({
df=consInputData()
a=summary(df)
c=grep(input$variablee1, colnames(df))
c=c[1]
valueBox(a[[1,c]],
"Minimum",
color = "red",
icon = icon("thumbs-o-down")
)
})
output$first1=renderValueBox({
df=consInputData()
a=summary(df)
c=grep(input$variablee1, colnames(df))
c=c[1]
valueBox(a[[2,c]],
"1st Quartile",
color = "maroon",
icon = icon("list-ol")
)
})
output$Median1=renderValueBox({
df=consInputData()
a=summary(df)
c=grep(input$variablee1, colnames(df))
c=c[1]
valueBox(a[[3,c]],
"Median",
color = "olive",
icon = icon("align-center")
)
})
output$Mean1=renderValueBox({
df=consInputData()
a=summary(df)
c=grep(input$variablee1, colnames(df))
c=c[1]
valueBox(a[[4,c]],
"Mean",
color = "teal",
icon = icon("align-center")
)
})
output$third1=renderValueBox({
df=consInputData()
a=summary(df)
c=grep(input$variablee1, colnames(df))
c=c[1]
valueBox(a[[5,c]],
"3rd Quartile",
color = "orange",
icon = icon("list-ol")
)
})
output$Max1=renderValueBox({
df=consInputData()
a=summary(df)
c=grep(input$variablee1, colnames(df))
c=c[1]
valueBox(a[[6,c]],
"Maximum",
color = "purple",
icon = icon("thumbs-o-up")
)
})
###############################################
#simple datatable of the data
output$rawDataView = renderDataTable({
newData = rawInputData();
if(is.null(newData))
return();
newData;
});
output$labelSelectUI = renderUI({
data = rawInputData();
return(selectInput("modelLabelUI","List of variables",colnames(data),colnames(data)[1]));
});
output$labelconsUI = renderUI({
data = consInputData();
return(selectInput("modelLabelUI","List of variables",colnames(data),colnames(data)[1]));
});
physicalData = reactive({
rd_phy = input$physicalInputFile
ht_phy = input$headUI_phy;
st_phy = input$sepaUI_phy;
qt_phy = input$quotUI_phy;
dec_phy = input$decUI_phy;
if(!is.null(rd_phy)) {
data = read.csv(rd_phy$datapath,header=ht_phy,sep=st_phy,quote=qt_phy,dec=dec_phy,row.names = 1);
} else {
return(NULL);
}
});
output$phy.data <- renderDataTable({
data = physicalData()
df.data = data.frame(data)
df.data
})
output$labelphyUI = renderUI({
data = physicalData();
return(selectInput("modelLabelUI","List of variables",colnames(data),colnames(data)[1]));
});
output$jitter <- renderUI({
if (identical(rawInputData()$input$product, '') || identical(rawInputData()$input$product,data.frame())) return(NULL)
checkboxInput('jitter', 'Jitter')
})
output$gdesc <- renderUI({
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("gdesc","Select descriptor variable:",items,multiple = T)
})
output$gfactor <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("gfactor", "Select a factor :",items)
})
plotAreac <- reactive({
data = rawInputData()
if(input$jitter){
ggplot(data=data, aes(x=data[,input$gfactor],y=data[,input$gdesc],fill=data[,input$gfactor])) +
geom_boxplot() +geom_jitter()+ theme(axis.text.x = element_text(angle = 45, hjust = 1)) + xlab("") + ylab("") + labs(fill = "")
}
else if(!input$jitter){
ggplot(data=data, aes(x=data[,input$gfactor],y=data[,input$gdesc],fill=data[,input$gfactor])) +
geom_boxplot() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) + xlab("") + ylab("") + labs(fill = "")
}
})
output$plotA <- renderPlot({
plotAreac()
})
output$dependent <- renderUI({
df <- rawInputData()
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("dependent","Select descriptor variable from:",items)
});
output$independents <- renderUI({
df <- rawInputData()
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
items =items*levels(items)
selectInput("independents","Select variable(s) from:",items,multiple=TRUE)
});
runRegression <- reactive({
lm(as.formula(paste(input$dependent," ~ ",paste(input$independents,collapse="+"))),data=rawInputData())
})
output$regTab <- renderPrint({
if(!is.null(input$independents)){
summary(runRegression())
} else {
print(data.frame(Warning="Please select Model Parameters."))
}
});
output$dependant <- renderUI({
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("dependant","Select response variable : ",items)
});
output$independants <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("independants", "Select explicative variables :",items,multiple = T)
});
runRegression <- reactive({
lm(as.formula(paste(input$dependent," ~ ",paste(input$independents,collapse="+"))),data=rawInputData())
})
output$regTab <- renderPrint({
if(!is.null(input$independents)){
summary(runRegression())
} else {
print(data.frame(Warning="Please select Model Parameters."))
}
})
output$dependant <- renderUI({
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("dependant","Select response variable : ",items)
})
output$independants <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("independants", "Select explicative variables :",items,multiple = T)
})
runAnova <- reactive({
aov(as.formula(paste(input$dependant," ~ ",paste(input$independants,collapse="*"))),data=rawInputData())
})
output$anov <- renderPrint({
summary.aov(runAnova())
})
get_factors <- reactive ({
factorstr <- as.character(formula(paste(input$dependant," ~ ",paste(input$independants,collapse="*"))))[3]
return(sub("\\s","",unlist(strsplit(factorstr,"[*+:]"))))
})
tukeyplot1 <- reactive({
dataset<-rawInputData()
aov.model<-runAnova()
out<-HSD.test(aov.model, get_factors())
# par(cex=1, mar=c(3,8,1,1))
# bar.group(out$groups,horiz=T,col="blue",
# xlim=c(0,max(out$means[,1]*1.2)),las=1)
data=data.frame(out$groups)
ggplot(data,aes(x=trt,y=means,fill=M,colour=M))+
geom_bar(stat = "identity",color="black")+
geom_text(aes(label=paste0(M),
y=means+0.3), size=5)+ ggtitle("tukey HSD")+ xlab("treatment groups")
})
output$tukeyplot <- renderPlot({
tukeyplot1()
})
output$downloadtuk <- downloadHandler(
filename = function() {
paste("tukeyPlot", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(tukeyplot1())
dev.off(which=dev.cur())}
)
output$tukey <- renderPrint({
dataset<-rawInputData()
aov.model<-runAnova()
out <-HSD.test(aov.model, get_factors(), " ")
print(out)
})
anovdd <-reactive(
as.data.frame(anova(runAnova()))
)
output$anov.d <-renderDataTable(
as.data.frame(anova(runAnova())), server = TRUE, filter = 'top', escape = FALSE, selection = 'none'
)
output$downloadtable <- downloadHandler(
filename = paste0("ANOVA_", Sys.Date(),".csv"),
content = function(file) {
write.csv(anovdd(),file,row.names=T)
}
)
output$variab <- renderUI({
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("variab","Select descriptor variable from:",items,multiple = T)
})
output$Vfactor <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("Vfactor", "Select a factor :",items)
})
plotBreac <- reactive({
df=rawInputData()
attach(df)
h <- list(
title = ""
)
l <- list(
title = "")
plot_ly(data=df,x=df[[input$variab]],color=df[[input$Vfactor]],colors = "Set2",type="box")%>%
layout(xaxis = h, yaxis = l)
})
output$plotB<-renderPlotly ({
plotBreac()
})
##Lines
output$vlines <- renderUI({
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("vlines","Select descriptor variable from:",items,multiple = T)
})
output$flines <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("flines", "Select a factor :",items)
})
output$plotD<-renderGvis ({
df=rawInputData()
attach(df)
return(gvisLineChart(df,xvar=input$flines,yvar=input$vlines, options=list(hAxis="{title:''}",vAxis="{title:''}",width=700,height=600)))
})
###Bubbles :
output$vbubble <- renderUI({
data<- rawInputData()
library(doBy)
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("vbubble","Select descriptor variable from:",items,multiple = T)
})
output$vbubble1 <- renderUI({
df=moy1()
df=data.frame(df,row.names = 1)
if (is.null(df)) return(NULL)
items=names(df)
names(items)=items
selectInput("vbubble1","Select descriptor variable from:",items,multiple = T)
})
output$f <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("f", "Choose your id var :",items)
})
output$f1 <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("f1", "Choose your size var :",items)
})
output$f2 <- renderUI({
df<- rawInputData()
if (is.null(df)) return(NULL)
#Let's only show factor columns
nums <- sapply(df, is.factor)
items=names(nums[nums])
names(items)=items
selectInput("f2", "Choose your color var :",items)
})
output$plotEr<-reactive ({
data=rawInputData()
return(gvisBubbleChart(data,idvar=input$f,xvar=input$vbubble,yvar=input$vbubble1,colorvar=input$f2,sizevar=input$f1,options=list(hAxis="{title:''}",vAxis="{title:''}",width=700,height=600)))
})
output$plotE<-renderGvis ({
data=rawInputData()
return(gvisBubbleChart(data,idvar=input$f,xvar=input$vbubble,yvar=input$vbubble1,colorvar=input$f2,sizevar=input$f1,options=list(hAxis="{title:''}",vAxis="{title:''}",width=700,height=600)))
})
output$plotF<-renderGvis({
data=rawInputData()
cdata=summaryBy(as.formula(paste(". ~ ",paste(input$product,"+",input$replication))),data=data,FUN=c(mean),na.rm=T,keep.names=T)
transform(cdata, time2 = paste(2010, cdata[,input$replication], sep = "")) -> df1
df1$time2 <- as.Date(df1$time2 ,"%Y-%m-%d")
return(gvisMotionChart(df1,idvar=df1[,input$product],timevar="time2",options=list(height=600,width=600,vAxes="[{title:''}",hAxes="[{title:''}")))
})
downloadPlotType <- reactive({
input$downloadPlotType
})
observe({
plotType <- input$downloadPlotType
plotTypePDF <- plotType == "pdf"
plotUnit <- ifelse(plotTypePDF, "inches", "pixels")
plotUnitDef <- ifelse(plotTypePDF, 9, 480)
updateNumericInput(
session,
inputId = "downloadPlotHeight",
label = sprintf("Height (%s)", plotUnit),
value = plotUnitDef)
updateNumericInput(
session,
inputId = "downloadPlotWidth",
label = sprintf("Width (%s)", plotUnit),
value = plotUnitDef)
})
# Get the download dimensions.
downloadPlotHeight <- reactive({
input$downloadPlotHeight
})
downloadPlotWidth <- reactive({
input$downloadPlotWidth
})
# Get the download file name.
downloadPlotFileName <- reactive({
input$downloadPlotFileName
})
# Include a downloadable file of the plot in the output list.
output$downloadPlot <- downloadHandler(
filename = function() {
paste(downloadPlotFileName(), downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(plotAreac())
dev.off(which=dev.cur())}
)
output$interval=renderUI({
Y=consInputData()
sliderInput("interval", "Interval of Consumers",
min = 1, max = ncol(Y), value = c(1,40),width = '60%')
})
output$cart<-renderPlot({cart() })
cart=eventReactive(input$actioncarto,{
Y=consInputData()
X=moy2()
hedo_senso=cbind.data.frame(Y[,input$interval[1]:input$interval[2]],X)
res.inter=PCA(hedo_senso ,scale.unit=T,quanti.sup = (ncol(Y[,input$interval[1]:input$interval[2]])+1):ncol(hedo_senso),graph=FALSE)
c1=fviz_pca_ind(res.inter, axes = c(1, 2), geom = c("point","text"),title="Plot of individuals" )
c3=fviz_pca_var(res.inter,invisible = "quanti.sup",geom=c( "arrow", "text"),title="Homogeniety of consumers")
c2=fviz_pca_var(res.inter,geom = c("arrow","text"),invisible ="var",title="Projection of sensory variables")
p=grid.arrange(c1,c2,c3,padding = unit(1, "line"),heights = 700)
p
})
output$downloadPlotinternal <- downloadHandler(
filename = function() {
paste("plot", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = 15, height = downloadPlotHeight())
plot(cart())
dev.off(which=dev.cur())}
)
################# External preference mapping
output$interval1=renderUI({
Y=consInputData()
sliderInput("interval1", "Interval of Consumers",
min = 1, max = ncol(Y), value = c(1,40))
})
output$modelch=renderUI({
selectInput("modelch", "Choose type of model :",choices = c("Vector model", "Circular model","Elliptic model",
"Quadratic model"))
})
output$Dimension_var<-renderUI({
selectInput("Dimension_var","Choose dimension reduction method:",choices = c("Principal Component Analysis", "Multiple Factor Analysis", "Canonical Analysis"))
})
output$Prediction_var<-renderUI({
selectInput("Prediction_var","Choose prediction model:",choices = c("Vector model", "Circular model","Elliptic model",
"Quadratic model", "Generalized Additive Model" , "Generalized Linear Model", "Bayesian model","Others"))
})
output$par_var<-renderUI({
if(input$Prediction_var != "Bayesian model"){
selectInput("par_var","If rejection of predictions outside liking scores:",choices = c("NO","YES"))}
else{selectInput("par_var","Choose Rejection of predictions outside of [0:10]:",choices = c("NO"))}
})
output$mod1<-renderUI({
if(input$Prediction_var == "Others"){
selectInput("mod1","Choose type of model:",choices = c("Polynomial regression","Generalized Additive Model" , "Generalized Linear Model", "Bayesian Model"))}
else return(NULL)
})
output$formula_lm1=renderUI({
if(input$Prediction_var == "Others"){
if(input$mod1 == "Polynomial regression"){
textInput("formula_lm1","Input formula for polynomial model :")}
else if(input$mod1 == "Generalized Additive Model"){
textInput("formula_lm1","Input formula for GAM :")}
else if(input$mod1 == "Generalized Linear Model"){
textInput("formula_lm1","Input formula for GLM :")}
else if(input$mod1 == "Bayesian Model"){
textInput("formula_lm1","Input formula for Bayesian model :")}}
})
output$interval2=renderUI({
Y=consInputData()
sliderInput("interval2", "Interval of Consumers",
min = 1, max = ncol(Y), value = c(1,40))
})
output$epmplot=renderPlot({cart13()})
cart13=eventReactive(input$actioncarto13,{
Y=consInputData()
X=moy2()
S=moy4()
nbpoints=50
if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
###############
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
#######################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
#################################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
#################################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
###############
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
#######################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
######################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{
drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
}
})
output$epmplot2=renderPlot({cart132()})
cart132=eventReactive(input$actioncarto13,{
Y=consInputData()
X=moy2()
S=moy4()
nbpoints=50
if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
###############
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
#######################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
#################################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
#################################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
##################
#################
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =TRUE, graph.map =FALSE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =TRUE, graph.map =FALSE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =TRUE, graph.map =FALSE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =TRUE, graph.map =FALSE,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
#######################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
######################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =T, graph.map =F,
graph.map.3D =FALSE )}
})
output$epmplot3=renderPlotly({cart133()})
cart133=eventReactive(input$actioncarto13,{
Y=consInputData()
X=moy2()
S=moy4()
nbpoints=50
if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )
}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
###############
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
#######################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
#################################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Vector model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Circular model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="YES")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
#################################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Vector model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Circular model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Elliptic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Quadratic model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Additive Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Generalized Linear Model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Bayesian model" && input$par_var=="NO")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
##################
#################
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =FALSE,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =FALSE,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =FALSE,
graph.map.3D =T)}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =FALSE,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Principal Component Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
#######################
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Multiple Factor Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
######################
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Polynomial regression")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Additive Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Generalized Linear Model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="YES"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
else if(input$Dimension_var=="Canonical Analysis" && input$Prediction_var=="Others" && input$par_var=="NO"
&&input$mod1=="Bayesian model")
{drawmap (Y[,input$interval2[1]:input$interval2[2]],X,S,axis=c(1,2),formula=input$formula_lm1,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =F, graph.map =F,
graph.map.3D =T )}
})
####### Results of models selection
output$Dimension_var1<-renderUI({
selectInput("Dimension_var1","Choose dimension reduction method:",choices = c("Principal Component Analysis", "Multiple Factor Analysis", "Canonical Analysis"))
})
output$Prediction_var1<-renderUI({
selectInput("Prediction_var1","Choose prediction model:",choices = c("Vector model", "Circular model","Elliptic model",
"Quadratic model", "Generalized Additive Model" , "Generalized Linear Model"))
})
output$aic<-renderUI({
selectInput("aic","Choose parameter of comparison:",choices = c("AIC","R2", "Fstat","nb-NA"))
})
output$model_selection=renderPlot({cart14()})
cart14=eventReactive(input$actioncarto14,{
Y=consInputData()
X=moy2()
S=moy4()
nbpoints=50
if(input$aic=="AIC" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Principal Component Analysis"){
map=map.with.pca(X)
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
aic.lm=extract.lm(x.lm,what=c("aic"))
aic.vect=extract.lm(x.vect,what=c("aic"))
aic.circ=extract.lm(x.circ,what=c("aic"))
aic.ellip=extract.lm(x.ellip,what=c("aic"))
aic.gam=extract.gam(x.gam,what=c("aic"))
aic.glm=extract.glm(x.glm,what=c("aic"))## ???
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(aic.lm,aic.vect,aic.circ,aic.ellip,aic.gam,aic.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
library(ggplot2)
bas=data.frame(x3,x2)
g2<-ggplot(bas,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("AIC")+ggtitle( "Comparison of models with AIC ")
g2<-g2+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g2
}
else if(input$aic=="AIC" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Multiple Factor Analysis"){
map=map.with.mfa(X,Y,axis=c(1,2))
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps)
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
aic.lm=extract.lm(x.lm,what=c("aic"))
aic.vect=extract.lm(x.vect,what=c("aic"))
aic.circ=extract.lm(x.circ,what=c("aic"))
aic.ellip=extract.lm(x.ellip,what=c("aic"))
aic.gam=extract.gam(x.gam,what=c("aic"))
aic.glm=extract.glm(x.glm,what=c("aic"))## ???
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(aic.lm,aic.vect,aic.circ,aic.ellip,aic.gam,aic.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
bas1=data.frame(x3,x2)
g21<-ggplot(bas1,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("AIC")+ggtitle( "Comparison of models with AIC ")
g21<-g21+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g21
}
else if(input$aic=="AIC" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Canonical Analysis"){
map=map.with.ca(X,S,Y)
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
aic.lm=extract.lm(x.lm,what=c("aic"))
aic.vect=extract.lm(x.vect,what=c("aic"))
aic.circ=extract.lm(x.circ,what=c("aic"))
aic.ellip=extract.lm(x.ellip,what=c("aic"))
aic.gam=extract.gam(x.gam,what=c("aic"))
aic.glm=extract.glm(x.glm,what=c("aic"))## ???
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(aic.lm,aic.vect,aic.circ,aic.ellip,aic.gam,aic.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
bas2=data.frame(x3,x2)
g23<-ggplot(bas2,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("AIC")+ggtitle( "Comparison of models with AIC ")
g23<-g23+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g23
}
else if(input$aic=="R2" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Principal Component Analysis"){
map=map.with.pca(X)
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
r2.lm=extract.lm(x.lm,what=c("rsquare"))
r2.vect=extract.lm(x.vect,what=c("rsquare"))
r2.circ=extract.lm(x.circ,what=c("rsquare"))
r2.ellip=extract.lm(x.ellip,what=c("rsquare"))
r2.gam=extract.gam(x.gam,what=c("rsquare"))
r2.glm=extract.gam(x.glm,what=c("rsquare"))
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(r2.lm,r2.vect,r2.circ,r2.ellip,r2.gam,r2.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
bas3=data.frame(x3,x2)
g22<-ggplot(bas3,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("R2")+ggtitle( "Comparison of models with R-squared ")
g22<-g22+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g22
}
else if(input$aic=="R2" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Multiple Factor Analysis"){
## If R2 and MFA
map=map.with.mfa(X,Y,axis=c(1,2))
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
r2.lm=extract.lm(x.lm,what=c("rsquare"))
r2.vect=extract.lm(x.vect,what=c("rsquare"))
r2.circ=extract.lm(x.circ,what=c("rsquare"))
r2.ellip=extract.lm(x.ellip,what=c("rsquare"))
r2.gam=extract.gam(x.gam,what=c("rsquare"))
r2.glm=extract.gam(x.glm,what=c("rsquare"))
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(r2.lm,r2.vect,r2.circ,r2.ellip,r2.gam,r2.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("Generalized Additive Model",ncol(Y)),rep("Generalized Linear Model",ncol(Y)))
bas4=data.frame(x3,x2)
g22<-ggplot(bas4,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("R2")+ggtitle( "Comparison of models with R-squared ")
g23<-g23+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g23
}
else if(input$aic=="R2" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Canonical Analysis"){
## If R2 and CA
map=map.with.ca(X,S,Y)
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
r2.lm=extract.lm(x.lm,what=c("rsquare"))
r2.vect=extract.lm(x.vect,what=c("rsquare"))
r2.circ=extract.lm(x.circ,what=c("rsquare"))
r2.ellip=extract.lm(x.ellip,what=c("rsquare"))
r2.gam=extract.gam(x.gam,what=c("rsquare"))
r2.glm=extract.gam(x.glm,what=c("rsquare"))
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(r2.lm,r2.vect,r2.circ,r2.ellip,r2.gam,r2.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
bas4=data.frame(x3,x2)
g24<-ggplot(bas4,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("R2")+ggtitle( "Comparison of models with R-squared ")
g24<-g24+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g24
}
else if(input$aic=="Fstat" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Principal Component Analysis"){
map=map.with.pca(X)
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
f.lm=extract.lm(x.lm,what=c("fstastic"))
f.vect=extract.lm(x.vect,what=c("fstastic"))
f.circ=extract.lm(x.circ,what=c("fstastic"))
f.ellip=extract.lm(x.ellip,what=c("fstastic"))
f.gam=extract.gam(x.gam,what=c("fstastic"))
f.glm=extract.glm(x.glm,what=c("fstastic"))
for (i in 1:length(f.glm)){ if (is.null(f.glm[[i]] )) f.glm[[i]]=0}
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(f.lm,f.vect,f.circ,f.ellip,f.gam,f.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
x2=unlist(x2)
bas5=data.frame(x3,x2)
g25<-ggplot(bas5,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("F-statistic")+ ggtitle( "Comparison of models with F-statistic ")
g25<-g25+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g25+ylim(min=0, max=100)
}
else if(input$aic=="Fstat" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Multiple Factor Analysis"){
map=map.with.mfa(X,Y,axis=c(1,2))
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
f.lm=extract.lm(x.lm,what=c("fstastic"))
f.vect=extract.lm(x.vect,what=c("fstastic"))
f.circ=extract.lm(x.circ,what=c("fstastic"))
f.ellip=extract.lm(x.ellip,what=c("fstastic"))
f.gam=extract.gam(x.gam,what=c("fstastic"))
f.glm=extract.glm(x.glm,what=c("fstastic"))
for (i in 1:length(f.glm)){ if (is.null(f.glm[[i]] )) {f.glm[[i]]=0} }
x2=c(f.lm,f.vect,f.circ,f.ellip,f.gam,f.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
x2=unlist(x2)
bas6=data.frame(x3,x2)
g26<-ggplot(bas6,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("F-statistic")+ ggtitle( "Comparison of models with F-statistic ")
g26<-g26+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g26+ylim(min=0, max=100)
}
else if(input$aic=="Fstat" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Canonical Analysis"){
## If fstat and CA
map=map.with.ca(X,S,Y)
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
attach(maps )
discretspace=discrete.function(map = maps)
x.lm=predict.scores.lm(Y = Y,discretspace = discretspace,map = maps)
x.vect=predict.scores.lm(Y = Y,formula="~F1+F2",discretspace = discretspace,map = maps)
x.circ=predict.scores.lm(Y = Y,formula="~ F1 + F2 + (F1*F1 + F2*F2)",discretspace = discretspace,map = maps)
x.ellip=predict.scores.lm(Y = Y,formula="~I(F1*F1)+I(F2*F2)",discretspace = discretspace,map = maps)
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,map = maps)
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,map = maps)
f.lm=extract.lm(x.lm,what=c("fstastic"))
f.vect=extract.lm(x.vect,what=c("fstastic"))
f.circ=extract.lm(x.circ,what=c("fstastic"))
f.ellip=extract.lm(x.ellip,what=c("fstastic"))
f.gam=extract.gam(x.gam,what=c("fstastic"))
f.glm=extract.glm(x.glm,what=c("fstastic"))
for (i in 1:length(f.glm)){ if (is.null(f.glm[[i]] )) {f.glm[[i]]=0} }
x2=c(f.lm,f.vect,f.circ,f.ellip,f.gam,f.glm)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
x2=unlist(x2)
bas7=data.frame(x3,x2)
g27<-ggplot(bas7,aes(x=x3,y=x2,fill=x3))+geom_boxplot()+xlab("Models")+
ylab("F-statistic")+ ggtitle( "Comparison of models with F-statistic ")
g27<-g27+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90),
plot.title = element_text(size=10, hjust = 0.5,face="bold"))
g27+ylim(min=0, max=100)
}
###########
else if(input$aic=="nb-NA" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Principal Component Analysis"){
map=map.with.pca(X) # pour mfa et ca tu changes cette commande
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
discretspace=discrete.function(map = maps)
x.quad=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", map = maps
,pred.na=TRUE)
nb.quad=x.quad$nb.NA
nb.quad=unlist(nb.quad)
nb.quad=as.data.frame(nb.quad)
#vect
x.vect=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~ F1 + F2", map = maps
,pred.na=TRUE)
nb.vect=x.vect$nb.NA
nb.vect=unlist(nb.vect)
nb.vect=as.data.frame(nb.vect)
#circ
x.circ=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", map = maps,
pred.na=TRUE)
nb.circ=x.circ$nb.NA
nb.circ=unlist(nb.circ)
nb.circ=as.data.frame(nb.circ)
#ellip
x.ellip=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~I(F1*F1)+I(F2*F2)", map = maps,
pred.na=TRUE)
nb.ellip=x.ellip$nb.NA
nb.ellip=unlist(nb.ellip)
nb.ellip=as.data.frame(nb.ellip)
# gam
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,
map = maps,
pred.na=TRUE)
nb.gam=x.gam$nb.NA
nb.gam=unlist(nb.gam)
nb.gam=as.data.frame(nb.gam)
# glm
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,
map = maps,
pred.na=TRUE)
nb.glm=x.glm$nb.NA
nb.glm=unlist(nb.glm)
nb.glm=as.data.frame(nb.glm)
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(nb.quad,nb.vect,nb.circ,nb.ellip,nb.gam,nb.glm)
x2=melt(x2)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
dt=cbind.data.frame(x1,x2[,1],x3)
colnames(dt)=c("consumer","occur.na","model")
gr2<-ggplot(dt,aes
(x=model,y=occur.na ,fill=model))+geom_boxplot()+xlab("Consumers")+ylab("Occurence of NA")+
ggtitle( "Comparison of nb-NA of prediction models from PCA ")
gr2+theme_bw()+ theme(plot.title = element_text(size=16, hjust = 0.5,face="bold"))
}
##################
else if(input$aic=="nb-NA" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Multiple Factor Analysis"){
map=map.with.mfa(X,Y,axis=c(1,2)) # pour mfa et ca tu changes cette commande
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
discretspace=discrete.function(map = maps)
x.quad=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", map = maps
,pred.na=TRUE)
nb.quad=x.quad$nb.NA
nb.quad=unlist(nb.quad)
nb.quad=as.data.frame(nb.quad)
#vect
x.vect=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~ F1 + F2", map = maps
,pred.na=TRUE)
nb.vect=x.vect$nb.NA
nb.vect=unlist(nb.vect)
nb.vect=as.data.frame(nb.vect)
#circ
x.circ=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", map = maps,
pred.na=TRUE)
nb.circ=x.circ$nb.NA
nb.circ=unlist(nb.circ)
nb.circ=as.data.frame(nb.circ)
#ellip
x.ellip=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~I(F1*F1)+I(F2*F2)", map = maps,
pred.na=TRUE)
nb.ellip=x.ellip$nb.NA
nb.ellip=unlist(nb.ellip)
nb.ellip=as.data.frame(nb.ellip)
# gam
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,
map = maps,
pred.na=TRUE)
nb.gam=x.gam$nb.NA
nb.gam=unlist(nb.gam)
nb.gam=as.data.frame(nb.gam)
# glm
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,
map = maps,
pred.na=TRUE)
nb.glm=x.glm$nb.NA
nb.glm=unlist(nb.glm)
nb.glm=as.data.frame(nb.glm)
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(nb.quad,nb.vect,nb.circ,nb.ellip,nb.gam,nb.glm)
x2=melt(x2)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
dt=cbind.data.frame(x1,x2[,1],x3)
colnames(dt)=c("consumer","occur.na","model")
gr2<-ggplot(dt,aes
(x=model,y=occur.na ,fill=model))+geom_boxplot()+xlab("Consumers")+ylab("Occurence of NA")+
ggtitle( "Comparison of nb-NA of prediction models from MFA ")
gr2+theme_bw()+ theme(plot.title = element_text(size=16, hjust = 0.5,face="bold"))
}
############
else if(input$aic=="nb-NA" && (input$Prediction_var1=="Vector model" ||input$Prediction_var1=="Circular model"||input$Prediction_var1=="Elliptic model"
||input$Prediction_var1=="Quadratic model"||input$Prediction_var1=="Generalized Additive Model"||input$Prediction_var1=="Generalized Linear Model") &&
input$Dimension_var1== "Canonical Analysis"){
map=map.with.ca(X,S,Y) # pour mfa et ca tu changes cette commande
maps=cbind.data.frame(map$F1,map$F2)
colnames(maps)=c("F1","F2")
discretspace=discrete.function(map = maps)
x.quad=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", map = maps
,pred.na=TRUE)
nb.quad=x.quad$nb.NA
nb.quad=unlist(nb.quad)
nb.quad=as.data.frame(nb.quad)
#vect
x.vect=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~ F1 + F2", map = maps
,pred.na=TRUE)
nb.vect=x.vect$nb.NA
nb.vect=unlist(nb.vect)
nb.vect=as.data.frame(nb.vect)
#circ
x.circ=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", map = maps,
pred.na=TRUE)
nb.circ=x.circ$nb.NA
nb.circ=unlist(nb.circ)
nb.circ=as.data.frame(nb.circ)
#ellip
x.ellip=predict.scores.lm(Y = Y,discretspace = discretspace,
formula="~I(F1*F1)+I(F2*F2)", map = maps,
pred.na=TRUE)
nb.ellip=x.ellip$nb.NA
nb.ellip=unlist(nb.ellip)
nb.ellip=as.data.frame(nb.ellip)
# gam
x.gam=predict.scores.gam(Y = Y,discretspace = discretspace,
map = maps,
pred.na=TRUE)
nb.gam=x.gam$nb.NA
nb.gam=unlist(nb.gam)
nb.gam=as.data.frame(nb.gam)
# glm
x.glm=predict.scores.glmulti(Y = Y,discretspace = discretspace,
map = maps,
pred.na=TRUE)
nb.glm=x.glm$nb.NA
nb.glm=unlist(nb.glm)
nb.glm=as.data.frame(nb.glm)
x1=c(1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y),1:ncol(Y))
x2=c(nb.quad,nb.vect,nb.circ,nb.ellip,nb.gam,nb.glm)
x2=melt(x2)
x3=c(rep("QR",ncol(Y)),rep("Vect",ncol(Y)),rep("Circ",ncol(Y)),rep("Ellip",ncol(Y)),rep("GAM",ncol(Y)),rep("GLM",ncol(Y)))
dt=cbind.data.frame(x1,x2[,1],x3)
colnames(dt)=c("consumer","occur.na","model")
gr2<-ggplot(dt,aes
(x=model,y=occur.na ,fill=model))+geom_boxplot()+xlab("Consumers")+ylab("Occurence of NA")+
ggtitle( "Comparison of nb-NA of prediction models from CA ")
gr2+theme_bw()+ theme(plot.title = element_text(size=16, hjust = 0.5,face="bold"))
}
})
output$downloadPlot_aic <- downloadHandler(
filename = function() {
paste("PLOT", downloadPlotType(), sep=".")
},
# The argument content below takes filename as a function
# and returns what's printed to it.
content = function(con) {
# Gets the name of the function to use from the
# downloadFileType reactive element. Example:
# returns function pdf() if downloadFileType == "pdf".
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(cart14())
dev.off(which=dev.cur())}
)
###### EPM smoothing
output$Dimension_var3<-renderUI({
selectInput("Dimension_var3","Choose dimension reduction method:",choices = c("Principal Component Analysis", "Multiple Factor Analysis", "Canonical Analysis"))
})
output$Prediction_var3<-renderUI({
selectInput("Prediction_var3","Choose prediction model:",choices = c("Vector model", "Circular model","Elliptic model",
"Quadratic model", "Generalized Additive Model" , "Generalized Linear Model", "Bayesian model","Others"))
})
output$mod2<-renderUI({
if(input$Prediction_var3 == "Others"){
selectInput("mod2","Choose type of model:",choices = c("Polynomial model","GAM" , "GLM", "Bayesian model"))}
else return(NULL)
})
output$formula_lm2=renderUI({
if(input$Prediction_var3 == "Others"){
if(input$mod2 == "Polynomial model"){
textInput("formula_lm2","Input formula for polynomial model :")}
else if(input$mod2 == "GAM"){
textInput("formula_lm2","Input formula for GAM :")}
else if(input$mod2 == "GLM"){
textInput("formula_lm2","Input formula for GLM :")}
else if(input$mod2 == "Bayesian model"){
textInput("formula_lm2","Input formula for Bayesian model :")}}
})
output$par_var3<-renderUI({
if(input$Prediction_var3 != "Bayesian model"){
selectInput("par_var3","If rejection of predictions outside of [0:10]:",choices = c("NO","YES"))}
else{selectInput("par_var3","If rejection of predictions outside of [0:10]:",choices = c("NO"))}
})
output$interval3=renderUI({
Y=consInputData()
sliderInput("interval3", "Interval of Consumers",
min = 1, max = ncol(Y), value = c(1,40))
})
output$smooth=renderPlot({cart15()})
cart15=eventReactive(input$actioncarto15,{
Y=consInputData()
X=moy2()
S=moy4()
nbpoints=50
if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Vector model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)
}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Circular model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Elliptic model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Quadratic model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Generalized Additive Model" && input$par_var3=="YES")
{par(mfrow = c(1,2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~s(F1,k=3)+s(F2,k=3)", dimredumethod=1,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Generalized Linear Model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=1,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
###############
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Vector model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Circular model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Elliptic model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Quadratic model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Generalized Additive Model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~s(F1,k=3)+s(F2,k=3)", dimredumethod=1,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Generalized Linear Model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=1,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Bayesian model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y,X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y,S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=1,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
################
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Vector model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Circular model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Elliptic model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Quadratic model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Generalized Additive Model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~s(F1,k=3)+s(F2,k=3)", dimredumethod=2,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Generalized Linear Model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=2,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
#######################
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Vector model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Circular model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Elliptic model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Quadratic model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Generalized Additive Model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~s(F1,k=3)+s(F2,k=3)", dimredumethod=2,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Generalized Linear Model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=2,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Bayesian model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=2,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
#################################
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Vector model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)
}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Circular model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Elliptic model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Quadratic model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Generalized Additive Model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~s(F1,k=3)+s(F2,k=3)", dimredumethod=3,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Generalized Linear Model" && input$par_var3=="YES")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=3,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
#################################
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Vector model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Circular model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~ F1 + F2 + (F1*F1 + F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~ F1 + F2 + (F1*F1 + F2*F2)", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Elliptic model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)
}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Quadratic model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Generalized Additive Model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~s(F1,k=3)+s(F2,k=3)",
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~s(F1,k=3)+s(F2,k=3)", dimredumethod=3,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Generalized Linear Model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=3,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Bayesian model" && input$par_var3=="NO")
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula="~I(F1*F1)+I(F2*F2)+F1*F2",
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula="~I(F1*F1)+I(F2*F2)+F1*F2", dimredumethod=3,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
#######################
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Polynomial regression" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Polynomial regression" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="GAM" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Generalized Additive Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Generalized Linear Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Generalized Linear Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Bayesian model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Principal Component Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Bayesian model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=1, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=1,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
#########################
#######################
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Polynomial regression" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Polynomial regression" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Generalized Additive Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Generalized Additive Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Generalized Linear Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Generalized Linear Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Bayesian model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Multiple Factor Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Bayesian model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=2, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=2,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
#########################
#######################
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Polynomial regression" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Polynomial regression" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=1, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=1, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Generalized Additive Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Generalized Additive Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=2, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=2, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Generalized Linear Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Generalized Linear Model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=3, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=3, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="NO" &&
input$mod2=="Bayesian model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =FALSE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=FALSE,dmap.loess=FALSE)}
else if(input$Dimension_var3=="Canonical Analysis" && input$Prediction_var3=="Others" && input$par_var3=="YES" &&
input$mod2=="Bayesian model" )
{par(mfrow = c(1, 2))
a=drawmap (Y[,input$interval3[1]:input$interval3[2]],X,S,axis=c(1,2),formula=input$formula_lm2,
dimredumethod=3, predmodel=4, nbpoints=50,
pred.na =TRUE, graph.pred =FALSE, graph.map =TRUE,
graph.map.3D =FALSE )
b=denoising.loess.global(X,Y[,input$interval3[1]:input$interval3[2]],S, axis=c(1,2), discretspace=discretspace,
formula=input$formula_lm2, dimredumethod=3,
predmodel=4, graphpred=FALSE, drawmap=TRUE,
pred.na=TRUE,dmap.loess=FALSE)}
})
##### comparison of maps stability
output$reduction<-renderUI({
selectInput("reduction","Choose dimension reduction method:",choices = c("Principal Component Analysis",
"Multiple Factor Analysis", "Canonical Analysis"))
})
output$Prediction_com<-renderUI({
selectInput("Prediction_com","Choose prediction model:",choices = c("Vector model", "Circular model","Elliptic model",
"Quadratic model", "Generalized Additive Model" , "Generalized Linear Model"))
})
output$formula_lm=renderUI({
textInput("formula_lm","Input formula for Polynomial and GLM models :")
})
output$formula_gam=renderUI({
textInput("formula_gam","Input formula for GAM :")
})
output$numm<-renderUI({
numericInput("numm","Introduce number of sampling n (at least 2):", 2,
min = 2, max = 100)
})
output$comparaison=renderPlot({cart16()})
cart16=eventReactive(input$actioncarto16,{
Y=consInputData()
X=moy2()
S=moy4()
# nbpoints=50
if(input$reduction=="Principal Component Analysis" && (input$Prediction_com=="Vector model" ||input$Prediction_com=="Circular model"||input$Prediction_com=="Elliptic model"
||input$Prediction_com=="Quadratic model"||input$Prediction_com=="Generalized Additive Model"||input$Prediction_com=="Generalized Linear Model"||input$Prediction_com=="Bayesian model")){
res= Dist_prob(Y,X,S,n=input$numm,axis=c(1,2),formula_lm=input$formula_lm,
formula_gam=input$formula_gam,dimredumethod=1,
nbpoints=50)
res=melt(res)
colnames(res)=c("prob", "var", "value")
gr<-ggplot(res,aes(x=prob,y=value,fill=prob))+geom_boxplot()+xlab("Methods")+
ylab("Difference Sum Squares of Preferences")
gr<-gr+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90))
gr
}
else if(input$reduction=="Multiple Factor Analysis" && (input$Prediction_com=="Vector model" ||input$Prediction_com=="Circular model"||input$Prediction_com=="Elliptic model"
||input$Prediction_com=="Quadratic model"||input$Prediction_com=="Generalized Additive Model"||input$Prediction_com=="Generalized Linear Model"||input$Prediction_com=="Bayesian model")){
res= Dist_prob(Y,X,S,n=input$numm,axis=c(1,2),formula_lm=input$formula_lm,
formula_gam=input$formula_gam,dimredumethod=2,
nbpoints=50)
res=melt(res)
colnames(res)=c("prob", "var", "value")
gr<-ggplot(res,aes(x=prob,y=value,fill=prob))+geom_boxplot()+xlab("Methods")+
ylab("Difference Sum Squares of Preferences")
gr<-gr+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90))
gr
}
else if(input$reduction=="Canonical Analysis" && (input$Prediction_com=="Vector model" ||input$Prediction_com=="Circular model"||input$Prediction_com=="Elliptic model"
||input$Prediction_com=="Quadratic model"||input$Prediction_com=="Generalized Additive Model"||input$Prediction_com=="Generalized Linear Model"||input$Prediction_com=="Bayesian model")){
res= Dist_prob(Y,X,S,n=input$numm,axis=c(1,2),formula_lm=input$formula_lm,
formula_gam=input$formula_gam,dimredumethod=3,
nbpoints=50)
res=melt(res)
colnames(res)=c("prob", "var", "value")
gr<-ggplot(res,aes(x=prob,y=value,fill=prob))+geom_boxplot()+xlab("Methods")+
ylab("Difference Sum Squares of Preferences")
gr<-gr+theme_bw()+theme(legend.position = "none",
axis.text.x = element_text(angle = 90))
gr
}
})
output$downloadPlot_comp <- downloadHandler(
filename = function() {
paste("comp", downloadPlotType(), sep=".")
},
# The argument content below takes filename as a function
# and returns what's printed to it.
content = function(con) {
# Gets the name of the function to use from the
# downloadFileType reactive element. Example:
# returns function pdf() if downloadFileType == "pdf".
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(cart16())
dev.off(which=dev.cur())}
)
output$dend<-renderPlot({
df=consInputData()
res.pca=PCA(t(df),graph = F)
res.hcpc=HCPC(res.pca,graph = F)
plot.HCPC(res.hcpc,choice="tree",new.plot=F)
})
clus1=reactive({
df=consInputData()
res.pca=PCA(t(df),graph = F)
res.hcpc=HCPC(res.pca,graph = F)
fviz_cluster(res.hcpc) + theme_minimal()+ ggtitle("Clustering Plot")
})
output$clus<-renderPlot({
clus1()
})
output$downloadclus <- downloadHandler(
filename = function() {
paste("Plot", downloadPlotType(), sep=".")
},
# The argument content below takes filename as a function
# and returns what's printed to it.
content = function(con) {
# Gets the name of the function to use from the
# downloadFileType reactive element. Example:
# returns function pdf() if downloadFileType == "pdf".
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(clus1())
dev.off(which=dev.cur())}
)
dend11<-reactive({
df=consInputData()
res.pca=PCA(t(df),graph = F)
res.hcpc=HCPC(res.pca,graph = F)
b=res.hcpc$call$t$tree
c=as.dendrogram(b)
nb=res.hcpc$call$t$nb.clust
d=c %>%
set("branches_k_color", k = nb) %>% set("branches_lwd", 0.7) %>%
set("labels_cex", 0.6) %>% set("labels_colors", k = nb) %>%
set("leaves_pch", 10) %>% set("leaves_cex", 0.5)%>%raise.dendrogram(0.015)%>%remove_nodes_nodePar
ggd1 <- as.ggdend(d)
ggplot(ggd1,labels=F)+ggtitle("Dendrogram")+scale_y_continuous(0,1.2)
plot.HCPC(res.hcpc,choice="tree",new.plot=F)
})
output$dend1<-renderPlot({
dend11()
})
output$downloaddend <- downloadHandler(
filename = function() {
paste("Plot", downloadPlotType(), sep=".")
},
content = function(con) {
plotFunction <- match.fun(downloadPlotType())
plotFunction(con, width = downloadPlotWidth(), height = downloadPlotHeight())
print(dend11())
dev.off(which=dev.cur())}
)
output$dend12<-renderPrint({
df=consInputData()
res.pca=PCA(t(df),graph = F)
res.hcpc=HCPC(res.pca,graph = F)
res.hcpc$desc.var
})
})
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.