R语言聚类循环报错修复：生成2-10类划分结果向量

Hey there! Let's break down why your code is throwing that error and get it working exactly how you need it.

What's Causing the Error? #

Your current code tries to assign the entire vector of cluster labels returned by cutree() to a single position somc[n] in a regular vector. Here's the core issue:

• cutree(hclust(dist(som_model$codes[[1]])), n) returns a vector where each element is the cluster label for one of your SOM codebook vectors (its length equals the number of rows in som_model$codes[[1]]).
• somc[n] is just a single element slot in a vector—you can't stuff a whole vector into one slot, hence the mismatch error.

The Fix: Use a List to Store Cluster Vectors #

You need a structure that can hold multiple vectors (one for each cluster count from 2 to 10). A list is perfect for this. Here's the corrected code:

# Initialize an empty list with 9 slots (for n=2 to 10)
somc <- vector("list", length = 9)
# Name the list elements for easy reference later
names(somc) <- as.character(2:10)

# Run the loop to populate each list element
for (n in 2:10) {
  # Assign the cluster vector to the corresponding list slot
  somc[[as.character(n)]] <- cutree(hclust(dist(som_model$codes[[1]])), n)
}

Even Cleaner: Use lapply Instead of a For Loop #

If you prefer more idiomatic R code, you can skip the for loop entirely with lapply:

# Generate all cluster vectors in one go
somc <- lapply(2:10, function(k) {
  cutree(hclust(dist(som_model$codes[[1]])), k)
})
# Add names to keep track of which element corresponds to how many clusters
names(somc) <- as.character(2:10)

How to Use the Result #

Now you can access each cluster assignment vector easily:

• somc[["2"]] gives you the 2-cluster partition
• somc[["5"]] gives you the 5-cluster partition
• And so on up to somc[["10"]]

内容的提问来源于stack exchange，提问作者Amri Muhaimin