错误：用 R 编写的 knn 函数中的外部函数调用（arg 6）中的 NA/NaN/Inf。有人可以纠正我吗？

Question

我使用的 knn 函数显示以下错误。有人可以帮帮我吗？

Error in knn(train = auction_data_train, test = auction_data_test, cl <- auction_data_train_labels,  : 

  NA/NaN/Inf in foreign function call (arg 6)

In addition: Warning messages:
1: In knn(train = auction_data_train, test = auction_data_test, cl <- auction_data_train_labels,  :
  NAs introduced by coercion
2: In knn(train = auction_data_train, test = auction_data_test, cl <- auction_data_train_labels,  :
  NAs introduced by coercion

我已经展示了我的数据结构和下面使用的代码。

> str(auction_data)
'data.frame':   69 obs. of  11 variables:
 $ Qty              : int  66 106 35 9 20 17 6 3 4 2 ...
 $ Unit.Price       : Factor w/ 15 levels "0-500","1000-1500",..: 12 12 13 15 15 3 9 9 8 14 ...
 $ New.Invoice.Date : int  1 1 1 1 1 1 1 1 1 1 ...
 $ New.Seller       : int  1 1 1 1 1 1 1 1 1 1 ...
 $ New.Seller.Branch: int  1 1 1 1 1 1 1 1 1 1 ...
 $ New.Location     : int  1 1 1 1 1 1 1 1 1 1 ...
 $ New.Buyer        : int  2 2 2 2 2 2 2 2 2 2 ...
 $ New.Make         : int  4 4 4 4 4 4 4 4 4 4 ...
 $ New.Model        : int  26 25 27 62 28 29 31 30 33 32 ...
 $ New.Category     : int  1 1 1 1 1 1 1 1 1 1 ...
 $ New.Sub.Category : int  2 2 2 2 2 2 2 2 2 2 ...
> auction_data_train<-auction_data[1:59, ]
> auction_data_test<-auction_data[60:69, ]
> auction_data_train_labels<-auction_data[1:59,2]
> auction_data_test_labels<-auction_data[60:69,2]
> library(class)
> auction_data_test_pred<-knn(train=auction_data_train, test=auction_data_test,cl<-auction_data_train_labels, k=8)

Answer 1

从训练数据和测试数据中排除标签。

auction_data_train<-auction_data[1:59, -2]
auction_data_test<-auction_data[60:69, -2] 

另外，在knn函数内部调用use cl = auction_data_train_labels。

注意：在函数调用中使用<- 运算符将创建一个新的（或覆盖现有的）对象。