Kohonen SOM 图在聚类图中显示观察结果。如何删除？

Question

我是运行 :

kmeansDat.t <- som_model$codes[[1]] %>% as.matrix


som_cluster <- cutree(hclust(dist(kmeansDat.t)), 5) %>% as.matrix
# plot these results:
plot(som_model, type="mapping", bgcol = pretty_palette[som_cluster], main = "Clusters") 
add.cluster.boundaries(som_model, som_cluster)

我的输出被黑色圆圈弄得乱七八糟，每个黑色圆圈似乎都代表每个节点中的观察数量。我怎样才能删除它们？

知道为什么会这样吗？

Answer 1

在plot.kohonen里面设置pchs=""解决问题：

library(kohonen)
library(magrittr)
# A dataset for testing the code
data(yeast)
X <- matrix(rnorm(100000), nrow=1000)
som_model <- som(X, somgrid(30, 30, "hexagonal"))
kmeansDat.t <- som_model$codes[[1]] %>% as.matrix
pretty_palette <- rainbow(5)   
som_cluster <- cutree(hclust(dist(kmeansDat.t)), 5) %>% as.matrix
# Plot Kohonen's map
plot(som_model, type="mapping", bgcol = pretty_palette[som_cluster], 
     main = "Clusters", pchs="") 
add.cluster.boundaries(som_model, som_cluster)

Kohonen SOM 图在聚类图中显示观察结果。如何删除？

Kohonen SOM plot is displaying observations within cluster plot. how to remove?

r

cluster-analysis

k-means

self-organizing-maps