Results
Outliers as given by randomForest
Cumulative Variance with Elbows
Hierarchical GMM Classifications
Hierarchical GMM Dendrogram
Restricting hGMM to \(K = 2\)
Here we are restricting hierarchical GMM to only go through on level. We are comparing the cluster results to the gaba
labels.
set.seed(314)
h2 <- hmc(sdat, maxDepth = 2, ccol = ccol, maxDim = 12)
h2lab <- viridis(max(h2$dat$labels$col))
K = 2 stacked means plot
stackM(h2, ccol = ccol, centered = TRUE, depth = 1)
Pairs plot colored by true gaba classification
cols <- c("black", "magenta")[gabaID$gaba+1]
acols <- alpha(cols, 0.35)
pairs(h2$dat$data, pch = 19, cex = 0.4, col = acols)
Pairs plot colored by hGMM cluster classification
acols2 <- alpha(h2lab[h2$dat$labels$col], 0.5)
pairs(h2$dat$data, pch = 19, cex = 0.4, col = acols2)
Permutation test for ARI
p0 <- mclust::adjustedRandIndex(pred, gaba)
perms <- foreach(i = 1:1.5e4, .combine = c) %dopar% {
set.seed(i*2)
mclust::adjustedRandIndex(sample(pred), gaba)
}
pv0 <- sum(c(perms,p0) >= p0)/length(perms)
hist(perms, xlim = c(min(perms), p0 + 0.25*p0),
main = "permutation test of ARI values", probability = TRUE)
#hist(perms, probability = TRUE)
abline(v = p0, col = 'red')
Summary Table
## truth
## pred FALSE TRUE
## FALSE 668 74
## TRUE 40 11
Misclassification Rate |
0.1437579 |
Accuracy |
0.8562421 |
Sensitivity |
0.1294118 |
Specificity |
0.9435028 |
Precision |
0.2156863 |
Recall |
0.1294118 |
ARI |
0.0731818 |
\(p\)-value for ARI |
0.013533 |
F1-score |
0.1617647 |
TP |
11 |
FP |
40 |
TN |
668 |
FN |
74 |