R 使用不同的列训练随机森林算法

我以前在这里问过这个问题，但还没有得到正确的答案。因此，这是一个更具可复制性的例子的另一次尝试

我有以下数据集：

train <- read.csv(url("http://s3.amazonaws.com/assets.datacamp.com/course/Kaggle/train.csv"))
test <- read.csv(url("http://s3.amazonaws.com/assets.datacamp.com/course/Kaggle/test.csv"))
train <- train[complete.cases(train), ]

列车
用于（列表中的R）{
莫菲特
#predict based on Pclass
fit <- randomForest(as.factor(Survived) ~ Pclass, data=train, importance=TRUE, ntree=2000)
Prediction <- predict(fit, test)
#fetch accuracy


#predict based on Pclass and Sex
fit <- randomForest(as.factor(Survived) ~ Pclass + Sex, data=train, importance=TRUE, ntree=2000)
Prediction <- predict(fit, test)
#fetch accuracy

list <- c(Pclass, Pclass + Sex)


for (R in list) {
  modfit <- paste0("won ~ ", R, ", data=training, method=\"rf\", prox=\"TRUE")
  modfit <- as.formula(modfit)
 train(modfit)
 }

Error in parse(text = x, keep.source = FALSE) : 
<text>:1:13: unexpected ','
1: won ~ Pclass,

for (R in list) {
 modfit <- paste0("won ~ ", R, "data=training, method=\"rf\", prox=\"TRUE")
 modfit <- as.formula(modfit)
 train(modfit)
 }