使用“ rf”方法的火车功能收到错误消息

Question

I tried the example posted on this site and followed the exact steps until the train function.

library(dplyr)

data_train <- read.csv("https://raw.githubusercontent.com/guru99-edu/R- 
    Programming/master/train.csv")

glimpse(data_train)

data_test <- read.csv("https://raw.githubusercontent.com/guru99-edu/R-    
    Programming/master/test.csv") 

glimpse(data_test)

library(randomForest)

library(caret)

library(e1071)

trControl <- trainControl(method = "cv",
    number = 10,
    search = "grid")

set.seed(1234)

rf_default <- train(Survived~., 
    data = data_train,
    method = "rf",
    metric = "Accuracy",
    trControl = trControl)

I used

R versions 3.5.1 and 3.6.1

Error in na.fail.default(list(Survived = c(0L, 1L, 1L, 1L, 0L, 0L, 0L, : missing values in object. However, there is no missing values in "Survived" variable.

Could someone tell me what's wrong please? I used R version 3.5.1, and tried on 3.6.1 as well. thank you

Answer 1

There are a few issues. The first being that you have NA s in there. You can either impute these or just omit them. For simplicity, I have omitted them.

Second, you need to use a factor for classification. set.seed(1234)

new_data<-na.omit(data_train)
as_tibble(new_data) %>% 
  mutate(Survived = as.factor(Survived)) -> new_data
rf_default <- train(Survived~., 
                    data = new_data,
                    method = "rf",
                    metric = "Accuracy",
                    trControl = trControl)