Java API implementing test time series string conversion by individual time series(rows) from CSV about sax

[balajikalluri]

@seninp

I'm looking for the equivalence of this R-implementation in Java.

`# classify the test data

labels_predicted = rep(-1, length(CBF[["labels_test"]]))

labels_test = CBF[["labels_test"]]

data_test = CBF[["data_test"]]

for (i in c(1:length(data_test[,1])))
{
print(paste(i))
series = data_test[i,]
bag = series_to_wordbag(series, w, p, a, "exact", 0.01)
cosines = cosine_sim(list("bag"=bag, "tfidf" = tfidf))
if (!any(is.na(cosines$cosines))) {
labels_predicted[i] = which(cosines$cosines == max(cosines$cosines))
}
}
`
Taken from your SAX-VSM R-page https://github.com/jMotif/jmotif-R

[seninp]

This functionality is not a part of this library and will not be. SAX lib is concerned with the discretization transform and the hot-sax algorithm. Sax-vsm lib implements the thing you want.