r - ggplot2:在 ROC 图上使用 scale_x_reverse

Question

我尝试使用 ggplot2 从 pRoc 包中的 plot.roc 函数重现 ROC 曲线。

library(mlbench)
library(caret)

data(Sonar)

set.seed(998)
fitControl <- trainControl(method = "repeatedcv",
                           number = 10,
                           repeats = 10,
                           ## Estimate class probabilities
                           classProbs = TRUE,
                           ## Evaluate performance using 
                           ## the following function
                           summaryFunction = twoClassSummary)

gbmGrid <-  expand.grid(interaction.depth = c(1, 5, 9),
                        n.trees = (1:30)*50,
                        shrinkage = 0.1,
                        n.minobsinnode = 20)

inTraining <- createDataPartition(Sonar$Class, p = .75, list = FALSE)
training <- Sonar[ inTraining,]
testing  <- Sonar[-inTraining,]


set.seed(825)
gbmFit <- train(Class ~ ., data = training,
                method = "gbm",
                trControl = fitControl,
                verbose = FALSE,
                tuneGrid = gbmGrid,
                ## Specify which metric to optimize
                metric = "ROC")
gbmFit

probs = predict(gbmFit, newdata = testing, type = "prob")

roc = roc(predictor = probs$M,
          response = testing$Class,
          levels = c('M','R'),
          percent = TRUE)
plot.roc(roc, print.auc = TRUE, col='red')


df = data.frame(Specificity=roc$specificities, Sensitivity=roc$sensitivities)
ggplot(data = df, aes(x = Specificity, y = Sensitivity))+
    geom_step(color='red', size=2, direction = "hv")+
    scale_x_reverse()+
    geom_abline(intercept = 100, slope = 1, color='grey')+
    annotate("text", x = 30, y = 20, label = paste0('AUC: ', round(roc$auc,1), '%'), size = 8)+
    ylab('Sensitivity (%)')+
    xlab('Specificity (%)')

plot.roc 产生:

当 ggplot2 产生:

scale_x_reverse() 似乎是问题所在，是否有任何其他方法可以反转 X 轴或更正该图？

Answer 1

您可以使用geom_path 代替geom_step:

ggplot(data = df, aes(x = Specificity, y = Sensitivity))+
  geom_path(colour = 'red', size = 2)+
  scale_x_reverse() +
  geom_abline(intercept = 100, slope = 1, color='grey')+
  annotate("text", x = 30, y = 20, label = paste0('AUC: ', round(roc$auc,1), '%'), size = 8)+
  ylab('Sensitivity (%)')+
  xlab('Specificity (%)')