caretEnsemble error: Error in FUN(X[[i]], ...) : { .... is not TRUE

Question

我一直在尝试将 2 个回归模型（glmnet 和 bagEarth）的预测叠加在一起，但我一直收到 "Error in FUN(X[[i]], ...) : { .... is not TRUE" 消息。根据我读过的内容，我发现这个问题源于重新采样索引，但由于我正在一起训练模型，所以我看不出如何解决这个问题。我已经能够使用随机数进行复制：

library(caret)
library(caretEnsemble)
rm(list=ls())

training <- as.data.frame(cbind(runif(24,1,100)
,runif(24,1,100)
,runif(24,1,100)
,runif(24,1,100)
,runif(24,1,100)
,runif(24,1,100)))

colnames(training) <- c("y", "x1", "x2", "x3", "x4", "x5")

set.seed(7)
ctrl <- trainControl(method = "cv", number = 3, returnResamp = "all", classProbs = FALSE, index = createMultiFolds(training$y, k = 3, times = 1))
model_list <- caretList(y~., data = training, trControl = ctrl, metric = "RMSE", methodList = c("glmnet", "bagEarth"))
train_ctrl <- trainControl(method = "cv", number = 3, classProbs = FALSE, savePredictions = TRUE, index = createMultiFolds(training$y, k = 3, times = 1))
glm_ensemble <- caretStack(model_list, method = "glm", metric = "RMSE", trControl = train_ctrl)

我知道我可能在某处遗漏了一个关键元素，欢迎任何意见。

谢谢，安东

Answer 1

进行了一些调试，错误来自一个名为 bestPreds 的函数。这是一个未导出的函数，它在 model_lists 中查找控制对象中保存的预测（全部或最终）。这个你没有在你的控制对象中设置。如果你添加这个，一切都会运行正常。我承认在这个地方出现错误消息会更好，而不仅仅是抛出错误。

ctrl <- trainControl(method = "cv", number = 3, returnResamp = "all", 
                     savePredictions = "final",  # needs to be final or all
                     classProbs = FALSE, index = createMultiFolds(training$y, k = 3, times = 1))

caretEnsemble error: Error in FUN(X[[i]], ...) : { .... is not TRUE

caretEnsemble error: Error in FUN(X[[i]], ...) : { .... is not TRUE

regression

r

ensemble-learning

r-caret