epclust/tests/testthat/test.computeMedoidsIndices.R

   1 context("computeMedoidsIndices")
   2
   3 test_that("serialization + getDataInFile retrieve original data / from matrix",
   4 {
   5     data_bin_file = "/tmp/epclust_test_m.bin"
   6     unlink(data_bin_file)
   7
   8     #dataset 200 lignes / 30 columns
   9     data_ascii = matrix(runif(200*30,-10,10),ncol=30)
  10     nbytes = 4 #lead to a precision of 1e-7 / 1e-8
  11     endian = "little"
  12
  13     #Simulate serialization in one single call
  14     binarize(data_ascii, data_bin_file, 500, ",", nbytes, endian)
  15     expect_equal(file.info(data_bin_file)$size, length(data_ascii)*nbytes+8)
  16     for (indices in list(c(1,3,5), 3:13, c(5,20,50), c(75,130:135), 196:200))
  17     {
  18         data_lines = getDataInFile(indices, data_bin_file, nbytes, endian)
  19         expect_equal(data_lines, data_ascii[indices,], tolerance=1e-6)
  20     }
  21     unlink(data_bin_file)
  22
  23     #...in several calls (last call complete, next call NULL)
  24     for (i in 1:20)
  25         binarize(data_ascii[((i-1)*10+1):(i*10),], data_bin_file, 20, ",", nbytes, endian)
  26     expect_equal(file.info(data_bin_file)$size, length(data_ascii)*nbytes+8)
  27     for (indices in list(c(1,3,5), 3:13, c(5,20,50), c(75,130:135), 196:200))
  28     {
  29         data_lines = getDataInFile(indices, data_bin_file, nbytes, endian)
  30         expect_equal(data_lines, data_ascii[indices,], tolerance=1e-6)
  31     }
  32     unlink(data_bin_file)
  33 })
  34
  35 TODO: test computeMedoids + filter
  36 #       #R-equivalent, requiring a matrix (thus potentially breaking "fit-in-memory" hope)
  37 #       mat_meds = medoids[,]
  38 #       mi = rep(NA,nb_series)
  39 #       for (i in 1:nb_series)
  40 #           mi[i] <- which.min( rowSums( sweep(mat_meds, 2, ref_series[i,], '-')^2 ) )
  41 #       rm(mat_meds); gc()
  42