Error: Metric Kappa not applicable for regression models

Question

我尝试使用KNN方法做预测，我已经将因子转换为字符，并且我的数据集包含数值和特征变量，缺失已被删除。这是我的 R 代码：

library(caret)
set.seed(142)

ctrl <- trainControl(method = "repeatedcv", number = 10, repeats = 5)   
train(LateStage ~ ., 
      data = finalP6, 
      method = "knn", 
      trControl = ctrl, 
      preProcess = c("range"),  
      tuneLength = 10,  
      metric = "Kappa")

我得到这个错误：

You are trying to do regression and your outcome only has two possible values Are you trying to do classification? 
If so, use a 2 level factor as your outcome column. 
Error: Metric Kappa not applicable for regression models

感谢您的帮助！

Answer 1

如果您提供数据示例并解释您尝试使用该模型做什么，将会有所帮助。

我假设您正在尝试使用 kNN 进行分类，如果您的自变量是因子，实际上就可以了。对于插入符号，您需要将因变量作为分类因素。

首先我生成了一些看起来像你的数据，（不要运行这个来解决你的问题）：

finalP6 = data.frame(LateStage = sample(0:1,1000,replace=TRUE),
matrix(runif(1000),ncol=10))

然后运行模型和你可以看到同样的错误：

library(caret)
set.seed(142)

ctrl <- trainControl(method = "repeatedcv", number = 10, repeats = 5)
train(LateStage ~ ., data = finalP6, method = "knn", 
trControl = ctrl, preProcess = c("range"),tuneLength = 10,
metric = "Kappa")

Error: Metric Kappa not applicable for regression models
In addition: Warning message:
In train.default(x, y, weights = w, ...) :
  You are trying to do regression and your outcome only has two possible values Are you trying to do classification? If so, use a 2 level factor as your outcome column.

您需要做的是：

finalP6<span class="math-container">$LateStage = factor(finalP6$</span>LateStage)
train(LateStage ~ ., data = finalP6, method = "knn", 
    trControl = ctrl, preProcess = c("range"),tuneLength = 10,
    metric = "Kappa")

k-Nearest Neighbors 

1000 samples
  10 predictor
   2 classes: '0', '1' 

Pre-processing: re-scaling to [0, 1] (10) 
Resampling: Cross-Validated (10 fold, repeated 5 times) 
Summary of sample sizes: 901, 900, 899, 900, 901, 900, ... 
Resampling results across tuning parameters:

  k   Accuracy   Kappa       
   5  0.4922998  -0.015433129
   7  0.4889058  -0.022246806
   9  0.5027024   0.005380261
  11  0.5074765   0.014693566
  13  0.5090744   0.017946500
  15  0.5100462   0.019756722
  17  0.5048402   0.009260903
  19  0.4868552  -0.027103397
  21  0.4842232  -0.032544799
  23  0.4798550  -0.041286084

Error: Metric Kappa not applicable for regression models

Error: Metric Kappa not applicable for regression models

machine-learning

knn