tests/testthat/tests_clust_methods.R

In reumandc/wsyn: Wavelet Approaches to Studies of Synchrony in Ecology and Other Fields

context("clust_methods")

#tests of set and get functions considered unnecessary

test_that("test print.clust",{
  set.seed(101)
  sig<-matrix(.8,5,5)
  diag(sig)<-1
  lents<-50
  #dat1<-t(mvtnorm::rmvnorm(lents,mean=rep(0,5),sigma=sig))
  #dat2<-t(mvtnorm::rmvnorm(lents,mean=rep(0,5),sigma=sig))
  dat1<-t(copy_rmvnorm(lents,mean=rep(0,5),sigma=sig)) #This copy function is in helper-utils.R. I made this change to avoid having to condition the performance of these tests on the availability of mvtnorm on the platform, since mvtnorm is in (and should be in) Suggests.
  dat2<-t(copy_rmvnorm(lents,mean=rep(0,5),sigma=sig))
  dat<-rbind(dat1,dat2)
  times<-1:lents
  dat<-cleandat(dat,times,clev=1)$cdat
  coords<-data.frame(Y=rep(0,10),X=1:10)
  method<-"coh.sig.fast"

  res<-clust(dat,times,coords,method,nsurrogs = 100)
  mod1val<-round(res$modres[[1]]$totQ,4)
  mod2val<-round(res$modres[[2]]$totQ,4)
  expect_equal(capture_output(print(res)),paste0("clust object:\ntimes, a length 50 numeric vector:\n1 2 3 4 5 ... 46 47 48 49 50 \nNumber of sampling locations: 10 \nmethodspecs:\nmethod=coh.sig.fast; tsrange=0 to Inf; nsurrogs=100; weighted=TRUE; sigthresh=0.95;\nscale.min=2; scale.max.input=NULL; sigma=1.05; f0=1\nadj has 88 of 90 off-diagonal entries differing from 0; values range from 0 to 0.990099 \nNumber of splitting steps done: 1 \nNumber of modules in final decomposition: 2 \nModularity values for each step: ",mod1val," ",mod2val," \nThe wmfs slot is: empty\nThe wpmfs slot is: empty"))

  res<-addwmfs(res)
  mod1val<-round(res$modres[[1]]$totQ,4)
  mod2val<-round(res$modres[[2]]$totQ,4)
  expect_equal(capture_output(print(res)),paste0("clust object:\ntimes, a length 50 numeric vector:\n1 2 3 4 5 ... 46 47 48 49 50 \nNumber of sampling locations: 10 \nmethodspecs:\nmethod=coh.sig.fast; tsrange=0 to Inf; nsurrogs=100; weighted=TRUE; sigthresh=0.95;\nscale.min=2; scale.max.input=NULL; sigma=1.05; f0=1\nadj has 88 of 90 off-diagonal entries differing from 0; values range from 0 to 0.990099 \nNumber of splitting steps done: 1 \nNumber of modules in final decomposition: 2 \nModularity values for each step: ",mod1val," ",mod2val," \nThe wmfs slot is: filled\nThe wpmfs slot is: empty"))
})

test_that("test summary.clust",{
  set.seed(101)
  sig<-matrix(.8,5,5)
  diag(sig)<-1
  lents<-50
  #dat1<-t(mvtnorm::rmvnorm(lents,mean=rep(0,5),sigma=sig))
  #dat2<-t(mvtnorm::rmvnorm(lents,mean=rep(0,5),sigma=sig))
  dat1<-t(copy_rmvnorm(lents,mean=rep(0,5),sigma=sig)) #This copy function is in helper-utils.R. I made this change to avoid having to condition the performance of these tests on the availability of mvtnorm on the platform, since mvtnorm is in (and should be in) Suggests.
  dat2<-t(copy_rmvnorm(lents,mean=rep(0,5),sigma=sig))
  dat<-rbind(dat1,dat2)
  times<-1:lents
  dat<-cleandat(dat,times,clev=1)$cdat
  coords<-data.frame(Y=rep(0,10),X=1:10)
  method<-"coh.sig.fast"
  res<-clust(dat,times,coords,method,nsurrogs = 100)
  
  h<-summary(res)
  expect_equal(class(h),c("summary_wsyn","list"))
  expect_equal(names(h),c("class","times_start","times_end","times_increment",
                          "sampling_locs",
                          "method","tsrange1","tsrange2","nsurrogs","weighted","sigthresh",
                          "scale.min","scale.max.input","sigma","f0",
                          "num_split_steps","num_final_modules","final_modularity",
                          "wmf_slot_is","wpmf_slot_is"))
  expect_equal(h[[1]],"clust")
  expect_equal(h[[2]],1)
  expect_equal(h[[3]],50)
  expect_equal(h[[4]],1)
  expect_equal(h[[5]],10)
  expect_equal(h[[6]],method)
  expect_equal(h[[7]],0)
  expect_equal(h[[8]],Inf)
  expect_equal(h[[9]],100)
  expect_equal(h[[10]],TRUE)
  expect_equal(h[[11]],.95)
  expect_equal(h[[12]],2)
  expect_equal(h[[13]],"NULL")
  expect_equal(h[[14]],1.05)
  expect_equal(h[[15]],1)
  expect_equal(h[[16]],1)
  expect_equal(h[[17]],2)
  expect_equal(h[[19]],"empty")
  expect_equal(h[[20]],"empty")
})