tests/testthat/test-d.spls.calval.R

In dual.spls: Dual Sparse Partial Least Squares Regression

test_that("d.spls.calval works", {

n <- 100
p <- 50
nondes <- 20
sigmaondes <- 0.5
data=d.spls.simulate(n=n,p=p,nondes=nondes,sigmaondes=sigmaondes)
X <- data$X
y <- data$y


ncells <- 5
Datatype <- d.spls.type(y,ncells=ncells)

pcal <- 70
#Error testing
expect_error(d.spls.calval(X,pcal=pcal,ncells=10))
expect_error(d.spls.calval(X,Datatype=Datatype,ncells=10))

L1=floor(0.7*length(which(Datatype==1)))
L2=floor(0.8*length(which(Datatype==2)))
L3=floor(0.6*length(which(Datatype==3)))
L4=floor(0.7*length(which(Datatype==4)))
L5=floor(0.8*length(which(Datatype==5)))

Listecal=c(L1,L2,L3,L4)

expect_error(d.spls.calval(X,Datatype=Datatype,Listecal=Listecal))

calval=d.spls.calval(X,y=y,pcal=pcal,Datatype=Datatype,ncells=5)
index.cal=calval$indcal
test0=which(index.cal==0)

# nb elts in calibration for each cell
Listecal=d.spls.listecal(Datatype,pcal)

# number of calibration points
  expect_equal(length(index.cal), sum(Listecal))


# content of calibration index vector"
  expect_equal(test0, integer())
})