test_that("assignMedoids behave as expected",
{
# Generate a gaussian mixture
- n = 999
- L = 7
- medoids = cbind( rep(0,L), rep(-5,L), rep(5,L) )
+ n <- 999
+ L <- 7
+ medoids <- cbind( rep(0,L), rep(-5,L), rep(5,L) )
# short series...
- series = t( rbind( MASS::mvrnorm(n/3, medoids[,1], diag(L)),
+ require("MASS", quietly=TRUE)
+ series <- t( rbind( MASS::mvrnorm(n/3, medoids[,1], diag(L)),
MASS::mvrnorm(n/3, medoids[,2], diag(L)),
MASS::mvrnorm(n/3, medoids[,3], diag(L)) ) )
# With high probability, medoids indices should resemble 1,1,1,...,2,2,2,...,3,3,3,...
- mi = epclust:::assignMedoids(medoids, series)
- mi_ref = rep(1:3, each=n/3)
+ mi <- assignMedoids(series, medoids)
+ mi_ref <- rep(1:3, each=n/3)
expect_lt( mean(mi != mi_ref), 0.01 )
-
- # Now with a random matrix, compare with (~trusted) R version
- series = matrix(runif(n*L, min=-7, max=7), nrow=L)
- mi = epclust:::assignMedoids(medoids, series)
- mi_ref = R_assignMedoids(medoids, series)
- expect_equal(mi, mi_ref)
})
-
-# R-equivalent of , requiring a matrix
-# (thus potentially breaking "fit-in-memory" hope)
-R_assignMedoids <- function(medoids, series)
-{
- nb_series = ncol(series) #series in columns
-
- mi = rep(NA,nb_series)
- for (i in 1:nb_series)
- mi[i] <- which.min( colSums( sweep(medoids, 1, series[,i], '-')^2 ) )
-
- mi
-}