Hi all,
Can you help me change my Kennard-Stone algorithm to faster one?
[The original code can run fast in matlab, but when I change matlab code to
R code, it is so slow.]
Since my code so crude and too many loops (changed from matlab code), it is
too slow. I hope that you can help to improve the performance.
Thanks.
kevin
################################################################################
#
##############################################################################
# Kennard-Stone algorithm for selection of samples
ksFun = function(x, N) {
# Initial the vector of minimum distance
dminmax = sample(0, N, replace = TRUE) # Default: FALSE
M = nrow(x)
samples = 1:M
# Initializes the matrix of distances
D = matrix(0, nrow = M, ncol = M)
for (i in 1:(M - 1)) {
xa = x[i, ]
for (j in (i + 1):M) {
xb = x[j, ]
# D: Upper Trianglar Matrix
# D[i, j] = Euclidean distance between object i and
j (j > i)
D[i, j] = max(svd(xa - xb)$d) # the largest
singular value
}
}
cat("for (i in 1:(M - 1)) done! \n")
maxD = apply(D, 2, max)
index_row = apply(D, 2, function(x) which(x == max(x))[1])
dummy = max(maxD)
index_column = which(maxD == dummy)
m = vector()
m[1] = index_row[index_column]
m[2] = index_column
dminmax[2] = D[m[1], m[2]]
for (i in 3:N) {
pool = setdiff(samples, m)
dmin = sample(0, M-i+1, replace = TRUE)
for (j in 1:(M-i+1)) {
indexa = pool[j]
d = sample(0, i-i, replace = TRUE)
for ( k in 1:(i-1)) {
indexb = m[k]
if (indexa < indexb) {
d[k] = D[indexa, indexb]
} else {
d[k] = D[indexb, indexa]
}
}
dmin[j] = min(d)
}
#cat("for (j in 1:(M -i+1)) done! \n")
dminmax[i] = max(dmin)
index = which(dmin == max(dmin))
m[i] = pool[index]
}
cat("for (i in 3:N) done! \n")
res = list(m = m, dminmax = dminmax)
return(res)
}
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
