context("clustering")
-#TODO: load some dataset ASCII CSV
-#data_bin_file <<- "/tmp/epclust_test.bin"
-#unlink(data_bin_file)
+#shorthand: map 1->1, 2->2, 3->3, 4->1, ..., 149->2, 150->3, ... (is base==3)
+I = function(i, base)
+ (i-1) %% base + 1
-test_that("computeClusters1 behave as expected",
+test_that("computeClusters1&2 behave as expected",
{
+ require("MASS", quietly=TRUE)
+ if (!require("clue", quietly=TRUE))
+ skip("'clue' package not available")
+ # 3 gaussian clusters, 300 items; and then 7 gaussian clusters, 490 items
+ n = 300
+ d = 5
+ K = 3
+ for (ndK in list( c(300,5,3), c(490,10,7) ))
+ {
+ n = ndK[1] ; d = ndK[2] ; K = ndK[3]
+ cs = n/K #cluster size
+ Id = diag(d)
+ coefs = sapply(1:K, function(i) MASS::mvrnorm(cs, c(rep(0,(i-1)),5,rep(0,d-i)), Id))
+ indices_medoids1 = computeClusters1(coefs, K, verbose=TRUE)
+ indices_medoids2 = computeClusters2(dist(coefs), K, verbose=TRUE)
+ # Get coefs assignments (to medoids)
+ assignment1 = sapply(seq_len(n), function(i)
+ which.min( colSums( sweep(coefs[,indices_medoids1],1,coefs[,i],'-')^2 ) ) )
+ assignment2 = sapply(seq_len(n), function(i)
+ which.min( colSums( sweep(coefs[,indices_medoids2],1,coefs[,i],'-')^2 ) ) )
+ for (i in 1:K)
+ {
+ expect_equal(sum(assignment1==i), cs, tolerance=5)
+ expect_equal(sum(assignment2==i), cs, tolerance=5)
+ }
+
+ costs_matrix1 = matrix(nrow=K,ncol=K)
+ costs_matrix2 = matrix(nrow=K,ncol=K)
+ for (i in 1:K)
+ {
+ for (j in 1:K)
+ {
+ # assign i (in result) to j (order 1,2,3)
+ costs_matrix1[i,j] = abs( mean(assignment1[((i-1)*cs+1):(i*cs)]) - j )
+ costs_matrix2[i,j] = abs( mean(assignment2[((i-1)*cs+1):(i*cs)]) - j )
+ }
+ }
+ permutation1 = as.integer( clue::solve_LSAP(costs_matrix1) )
+ permutation2 = as.integer( clue::solve_LSAP(costs_matrix2) )
+ for (i in 1:K)
+ {
+ expect_equal(
+ mean(assignment1[((i-1)*cs+1):(i*cs)]), permutation1[i], tolerance=0.05)
+ expect_equal(
+ mean(assignment2[((i-1)*cs+1):(i*cs)]), permutation2[i], tolerance=0.05)
+ }
+ }
})
test_that("computeSynchrones behave as expected",
{
+ n = 300
+ x = seq(0,9.5,0.1)
+ L = length(x) #96 1/4h
+ K = 3
+ s1 = cos(x)
+ s2 = sin(x)
+ s3 = c( s1[1:(L%/%2)] , s2[(L%/%2+1):L] )
+ #sum((s1-s2)^2) == 96
+ #sum((s1-s3)^2) == 58
+ #sum((s2-s3)^2) == 38
+ s = list(s1, s2, s3)
+ series = matrix(nrow=L, ncol=n)
+ for (i in seq_len(n))
+ series[,i] = s[[I(i,K)]] + rnorm(L,sd=0.01)
+ getRefSeries = function(indices) {
+ indices = indices[indices <= n]
+ if (length(indices)>0) series[,indices] else NULL
+ }
+ synchrones = computeSynchrones(bigmemory::as.big.matrix(cbind(s1,s2,s3)), getRefSeries,
+ n, 100, verbose=TRUE, parll=FALSE)
+ expect_equal(dim(synchrones), c(L,K))
+ for (i in 1:K)
+ expect_equal(synchrones[,i], s[[i]], tolerance=0.01)
})
-test_that("computeClusters2 behave as expected",
+# NOTE: medoids can be a big.matrix
+computeDistortion = function(series, medoids)
+{
+ n = nrow(series) ; L = ncol(series)
+ distortion = 0.
+ if (bigmemory::is.big.matrix(medoids))
+ medoids = medoids[,]
+ for (i in seq_len(n))
+ distortion = distortion + min( colSums( sweep(medoids,1,series[,i],'-')^2 ) / L )
+ distortion / n
+}
+
+test_that("clusteringTask1 behave as expected",
{
+ n = 900
+ x = seq(0,9.5,0.1)
+ L = length(x) #96 1/4h
+ K1 = 60
+ s = lapply( seq_len(K1), function(i) x^(1+i/30)*cos(x+i) )
+ series = matrix(nrow=n, ncol=L)
+ for (i in seq_len(n))
+ series[,i] = s[[I(i,K1)]] + rnorm(L,sd=0.01)
+ getSeries = function(indices) {
+ indices = indices[indices <= n]
+ if (length(indices)>0) series[,indices] else NULL
+ }
+ wf = "haar"
+ ctype = "absolute"
+ getContribs = function(indices) curvesToContribs(series[,indices],wf,ctype)
+ indices1 = clusteringTask1(1:n, getContribs, K1, 75, verbose=TRUE, parll=FALSE)
+ medoids_K1 = getSeries(indices1)
+ expect_equal(dim(medoids_K1), c(L,K1))
+ # Not easy to evaluate result: at least we expect it to be better than random selection of
+ # medoids within initial series
+ distorGood = computeDistortion(series, medoids_K1)
+ for (i in 1:3)
+ expect_lte( distorGood, computeDistortion(series,series[,sample(1:n, K1)]) )
})
-test_that("clusteringTask + computeClusters2 behave as expected",
+test_that("clusteringTask2 behave as expected",
{
+ n = 900
+ x = seq(0,9.5,0.1)
+ L = length(x) #96 1/4h
+ K1 = 60
+ K2 = 3
+ #for (i in 1:60) {plot(x^(1+i/30)*cos(x+i),type="l",col=i,ylim=c(-50,50)); par(new=TRUE)}
+ s = lapply( seq_len(K1), function(i) x^(1+i/30)*cos(x+i) )
+ series = matrix(nrow=n, ncol=L)
+ for (i in seq_len(n))
+ series[i,] = s[[I(i,K1)]] + rnorm(L,sd=0.01)
+ getRefSeries = function(indices) {
+ indices = indices[indices <= n]
+ if (length(indices)>0) series[,indices] else NULL
+ }
+ # Artificially simulate 60 medoids - perfect situation, all equal to one of the refs
+ medoids_K1 = bigmemory::as.big.matrix( sapply( 1:K1, function(i) s[[I(i,K1)]] ) )
+ medoids_K2 = clusteringTask2(medoids_K1, K2, getRefSeries, n, 75, verbose=TRUE, parll=FALSE)
+ expect_equal(dim(medoids_K2), c(L,K2))
+ # Not easy to evaluate result: at least we expect it to be better than random selection of
+ # medoids within 1...K1 (among references)
+ distorGood = computeDistortion(series, medoids_K2)
+ for (i in 1:3)
+ expect_lte( distorGood, computeDistortion(series,medoids_K1[,sample(1:K1, K2)]) )
})
+
+#NOTE: rather redundant test
+#test_that("clusteringTask1 + clusteringTask2 behave as expected",
+#{
+# n = 900
+# x = seq(0,9.5,0.1)
+# L = length(x) #96 1/4h
+# K1 = 60
+# K2 = 3
+# s = lapply( seq_len(K1), function(i) x^(1+i/30)*cos(x+i) )
+# series = matrix(nrow=n, ncol=L)
+# for (i in seq_len(n))
+# series[i,] = s[[I(i,K1)]] + rnorm(L,sd=0.01)
+# getSeries = function(indices) {
+# indices = indices[indices <= n]
+# if (length(indices)>0) series[indices,] else NULL
+# }
+# wf = "haar"
+# ctype = "absolute"
+# getContribs = function(indices) curvesToContribs(series[indices,],wf,ctype)
+# require("bigmemory", quietly=TRUE)
+# indices1 = clusteringTask1(1:n, getContribs, K1, 75, verbose=TRUE, parll=FALSE)
+# medoids_K1 = bigmemory::as.big.matrix( getSeries(indices1) )
+# medoids_K2 = clusteringTask2(medoids_K1, K2, getSeries, n, 120, verbose=TRUE, parll=FALSE)
+#
+# expect_equal(dim(medoids_K1), c(K1,L))
+# expect_equal(dim(medoids_K2), c(K2,L))
+# # Not easy to evaluate result: at least we expect it to be better than random selection of
+# # medoids within 1...K1 (among references)
+# distorGood = computeDistortion(series, medoids_K2)
+# for (i in 1:3)
+# expect_lte( distorGood, computeDistortion(series,medoids_K1[sample(1:K1, K2),]) )
+#})
projects / epclust.git / blobdiff