如何改进 R 中的混淆矩阵？

Question

使用 R 中的鸢尾花数据集，我编写了一个函数来绘制混淆矩阵。

library(e1071)
library(caTools)
library(caret)

iris$spl = sample.split(iris, SplitRatio = 0.1)
train <- subset(iris, iris$spl == TRUE)
test <- subset(iris, iris$spl == FALSE)

iris.nb <- naiveBayes(Species ~ ., data = train)

nb_train_predict <- predict(iris.nb, test[ , names(test) != "Species"])

cfm <- confusionMatrix(nb_train_predict, test$Species)
cfm

#ggplot confusion matrix
library(ggplot2)
library(scales)

ggplotConfusionMatrix <- function(m){
  mytitle <- paste("Accuracy", percent_format()(m$overall[1]),
                   "Kappa", percent_format()(m$overall[2]))
  p <-
    ggplot(data = as.data.frame(m$table) ,
           aes(x = Reference, y = Prediction)) +
    geom_tile(aes(fill = log(Freq)), colour = "white") +
    scale_fill_gradient(low = "white", high = "steelblue") +
    geom_text(aes(x = Reference, y = Prediction, label = Freq)) +
    theme(legend.position = "none") +
    ggtitle(mytitle)
  return(p)
}

ggplotConfusionMatrix(cfm)

我的问题是：如何从矩阵中删除“0”？（我想要没有任何文字的灰色方块）

Answer 1

您可以为标签创建单独的列。对于 0 频率，将它们设为空白。

library(ggplot2)
library(scales)

ggplotConfusionMatrix <- function(m){
  mytitle <- paste("Accuracy", percent_format()(m$overall[1]),
                   "Kappa", percent_format()(m$overall[2]))
  dat <- as.data.frame(m$table)
  dat$lab <- ifelse(dat$Freq == 0, '', dat$Freq)
  p <-
    ggplot(data = dat ,
           aes(x = Reference, y = Prediction)) +
    geom_tile(aes(fill = log(Freq)), colour = "white") +
    scale_fill_gradient(low = "white", high = "steelblue") +
    geom_text(aes(x = Reference, y = Prediction, label = lab)) +
    theme(legend.position = "none") +
    ggtitle(mytitle)
  return(p)
}

ggplotConfusionMatrix(cfm)

如何改进 R 中的混淆矩阵？

how can I improve this confusion matrix in R?

r

knn

ggplot2

confusion-matrix