You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to calculate k-means and PCA from a distribution of rotation data without time series information. However, what I expected would work (constructing a qts_sample series where each sample is one point long, and specifying no time-warping for kmeans) instead led to errors. Is there some other approach I should take, or am I misunderstanding what is necessary to complete k-means/PCA/distance matrix in rotation space?
If helpful, here's a reprex:
library(squat)
# works
working_sample <- as_qts_sample(lapply(vespa64$igp[1:20], function(m) {m[1:10,]}))
kmeans(working_sample, 2, warping_class="none")
# does not work
failing_sample <- as_qts_sample(lapply(vespa64$igp[1:20], function(m) {m[1,]}))
kmeans(failing_sample, 2, warping_class="none")
I'm trying to calculate k-means and PCA from a distribution of rotation data without time series information. However, what I expected would work (constructing a qts_sample series where each sample is one point long, and specifying no time-warping for kmeans) instead led to errors. Is there some other approach I should take, or am I misunderstanding what is necessary to complete k-means/PCA/distance matrix in rotation space?
If helpful, here's a reprex:
Thanks!
The text was updated successfully, but these errors were encountered: