tests/testthat/test_PAA.R

In jmotif: Time Series Analysis Toolkit Based on Symbolic Aggregate Discretization, i.e. SAX

y <- c(-1, -2, -1, 0, 2, 1, 1, 0)
v <- c(-1.375, 0.75, 0.625)

test_that("points to points via PAA", {
  expect_equal(paa(y, 3), v)
  expect_equal(paa(y, 8), y)
  expect_equal(paa(y, 2), c(-1, 1))
})


dat <- c(-0.9796808, -0.8622706, -0.6123005, 0.8496459, 1.739691, 1.588194, 1.095829,
         0.5277147, 0.4709033, -0.2865819, 0.0921607, -0.2865819, -0.9039323, -1.195564,
         -1.237226)
dat_paa <- c(-0.9405441, -0.6956239, 1.146328, 1.638693, 0.9064573, 0.4898404,
             -0.1603344, -0.1603344, -1.001143, -1.223339)
test_that("points to points via PAA_CPP #2", {
  expect_equal(paa(dat, 10), dat_paa, tolerance = 1e-6)
})

dat <- c(-1.289433, -0.9992189, -0.5253246, -0.06612478, -0.2791935, 0.08816637,
         -0.06612478, 0.595123, 0.8926845, 0.8228861, 1.741286, 1.770675, -0.2791935,
         -1.197593, -1.208614)
dat_paa <- c(-1.192695, -0.6832894, -0.1371477, -0.03428692, 0.1542912, 0.7934974,
             1.129019, 1.760878, -0.5853268, -1.20494)
test_that("points to points via PAA_CPP #3", {
  expect_equal(paa(dat, 10), dat_paa, tolerance = 1e-6)
})