R语言kamila聚类包报错：Input datasets must be dataframes（单连续变量）

Hey, I’ve run into this exact quirk with kamila before! Let me break down what’s happening and how to fix it quickly:

When you pull just one continuous variable using data[, conInd], R automatically simplifies the result to a vector instead of keeping it as a data frame. But the kamila() function strictly expects the continuous variable input to be a data frame, which is why you hit that error. When you use two variables, the extraction stays as a data frame by default, so everything works smoothly.

Here’s the straightforward fix: #

Add the drop=FALSE parameter when extracting your single continuous variable—this forces R to preserve the data frame structure instead of collapsing it to a vector:

library(kamila)
data <- read.csv("mixed.csv", header=FALSE, sep=";")
conInd <- 9
# This line is the key change: drop=FALSE keeps the result as a data frame
conVars <- data[, conInd, drop=FALSE]
conVars <- data.frame(scale(conVars))

catVarsFac <- data[, c(1,2,3,4,5,6,7,8)]
catVarsFac[] <- lapply(catVarsFac, factor)

kamRes <- kamila(conVars, catVarsFac, numClust=5, numInit=10,
                 calcNumClust = "ps", numPredStrCvRun = 10, predStrThresh = 0.5)

Quick check to confirm it works: #

Before running kamila(), you can verify conVars is a data frame with this quick base R command:

is.data.frame(conVars)

It should return TRUE after making the change.

If you’re comfortable with the tidyverse, another option is using dplyr::select() to pull the column—it always returns a data frame even for single columns—but the drop=FALSE trick is simpler if you’re sticking to base R.

内容的提问来源于stack exchange，提问作者Emrah BILGIC