make no longer a draft
diff --git a/site/src/site/blog/img/matrixWhiskyCorrHeatmap.png b/site/src/site/blog/img/matrixWhiskyCorrHeatmap.png
index 556d134..b5d535f 100644
--- a/site/src/site/blog/img/matrixWhiskyCorrHeatmap.png
+++ b/site/src/site/blog/img/matrixWhiskyCorrHeatmap.png
Binary files differ
diff --git a/site/src/site/blog/whisky-revisited.adoc b/site/src/site/blog/whisky-revisited.adoc
index c1a1286..3c78aa0 100644
--- a/site/src/site/blog/whisky-revisited.adoc
+++ b/site/src/site/blog/whisky-revisited.adoc
@@ -1,7 +1,6 @@
= Whisky flavor profiles revisited
Paul King
:revdate: 2025-04-17T22:30:00+00:00
-:draft: true
:keywords: whisky, whiskey, groovy, kmeans, clustering, apache echarts
:description: This post looks at using the Underdog data science library.
@@ -258,7 +257,7 @@
[source,groovy]
----
(summary.columns - 'Cluster').each { c ->
- summary[c] = summary[c](Double, Double) {it.round(3) }
+ summary[c] = summary[c](Double, Double) { it.round(3) }
}
println summary
----
@@ -274,7 +273,7 @@
2 | 1.5 | 2.455 | 1.114 | 0.227 | 0.114 | 1.114 | 1.114 | 0.591 | 1.25 | 1.818 | 1.773 | 1.977 |
----
-Looking at the centroids is one way to understand how the whiskies have been grouped.
+Looking at the centroids is one way to try to understand how the whiskies have been grouped.
But, it's very hard to visualize 12 dimensional data, so instead,
let's project our data onto 2 dimensions using PCA and store those projections back into the dataframe:
@@ -537,7 +536,7 @@
[source,groovy]
----
def corr = [size<..0, 0..<size].combinations().collect { i, j ->
- Correlation.cor(data*.getAt(j), data*.getAt(i)) * 100 as int
+ Correlation.cor(data*.getAt(j), data*.getAt(i)).round(2)
}
def corrMatrix = Matrix.builder().data(X: 0..<corr.size(), Heat: corr)
@@ -564,9 +563,9 @@
== Conclusion
-We have looked at how to use Underdog and Matrix.
+We have looked at how to use Underdog and Matrix to classify whisky flavor profiles.
.Update history
****
-*19/Apr/2025*: Initial version +
+*02/May/2025*: Initial version +
****
