在分层k-Fold的情况下如何计算混淆矩阵的总和答案

【问题标题】：How to calculate sum of confusion matrix in the case of stratified k- Fold在分层k-Fold的情况下如何计算混淆矩阵的总和
【发布时间】：2019-12-02 23:42:04
【问题描述】：

我是机器学习的新手。我在我的数据集上应用了分层的 k 折。我将如何获得混淆矩阵的总和。

下面是我的代码

library(MASS)

cv_lda <- lapply(folds, function(x) { # start of function
  # in the next two lines we will separate the Training set into it's 10 pieces
  training_fold = ForwardPlayers[-x, ] # training fold =  training set minus (-) it's sub test fold
  test_fold = ForwardPlayers[x, ] # here we describe the test fold individually
  # now apply (train) the classifer on the training_fold
  classifier = lda(Rating ~ .,training_fold)
  # next step in the loop, we calculate the predictions and cm and we equate the accuracy
  # note we are training on training_fold and testing its accuracy on the test_fold
  y_pred = predict(classifier, newdata = test_fold[-1]) 

  cm = table(test_fold$Rating, y_pred$class)
  #accuracy <- sum(diag(cm))/sum(cm)

  return(cm)
})

【问题讨论】：

标签： r machine-learning confusion-matrix

【解决方案1】：

你试过了吗？

addmargins()

-> 或任何边距子参数

作为论据？

【讨论】：

【解决方案2】：

如果您的意思是矩阵中的全部数据的总和来衡量 P+N 的准确度指标，您可以试试这个： sum(sum(confusion_matrix))

【讨论】：

请在答案中包含额外的详细信息，并尝试始终将代码包含在“代码示例”块中。