应用统计II项目R语言随机森林代码报错：参数长度不一致求助

问题描述 #

我是一名R语言新手，正在完成应用统计II（Applied Stats II）项目。运行代码时反复出现以下错误：

[1] "Confusion Matrix: TRAINING set based on random forest model built using 2 trees"
Error in table(train.data$heart_data, train.data.predict): all arguments must have the same length
Traceback:

1. table(train.data$heart_data, train.data.predict)
2. stop("all arguments must have the same length")

运行代码 #

set.seed(6522048)
library(randomForest)
model_rf1 <- randomForest(target ~ age+sex+cp+trestbps+chol+restecg+exang+ca, 
                          data = train.data, ntree = 2)
print("======================================================================================================================")
print('Confusion Matrix: TRAINING set based on random forest model built using 2 trees')
train.data.predict <- predict(model_rf1, train.data, type = "class")

conf.matrix <- table(train.data$heart_data, train.data.predict)[,c('YES','NO')]
rownames(conf.matrix) <- paste("Actual", rownames(conf.matrix), sep = ": ")
colnames(conf.matrix) <- paste("Prediction", colnames(conf.matrix), sep = ": ")

format(conf.matrix,justify="centre",digit=2)

print("=====================================================================================================================")
print('Confusion Matrix: TESTING set based on random forest model built using 2 trees')
test.data.predict <- predict(model_rf1, test.data, type = "class")

conf.matrix <- table(test.data$heart_data, test.data.predict)[,c('YES','NO')]
(conf.matrix) <- paste("Actual", rownames(conf.matrix), sep = ": ")
colnames(conf.matrix) <- paste("Prediction", colnames(conf.matrix), sep = ": ")

format(conf.matrix,justify="centre",digit=2)

错误原因与修复方案 #

核心问题 #

1. 因变量名称不匹配：模型训练时用的因变量是target，但生成混淆矩阵时却调用了train.data$heart_data，二者要么不是同一变量，要么heart_data的长度和预测结果train.data.predict不匹配，导致table()函数报错。
2. 语法错误：测试集部分的(conf.matrix) <- paste(...)多了一对括号，会将字符串直接赋值给conf.matrix，破坏后续矩阵操作逻辑。

修复步骤 #

1. 统一因变量名称
   保持模型与混淆矩阵的因变量一致：

   • 如果模型中的因变量确实是target，修改混淆矩阵的table行：

     # 训练集
     conf.matrix <- table(train.data$target, train.data.predict)[,c('YES','NO')]
     # 测试集
     conf.matrix <- table(test.data$target, test.data.predict)[,c('YES','NO')]
     

   • 如果你的因变量实际是heart_data，则修改模型公式为heart_data ~ age+sex+cp+trestbps+chol+restecg+exang+ca。

2. 修复测试集语法错误
   将测试集部分的(conf.matrix) <- paste(...)修正为rownames(conf.matrix) <- paste(...)，与训练集逻辑保持一致：

   rownames(conf.matrix) <- paste("Actual", rownames(conf.matrix), sep = ": ")
   

3. 可选：提前验证长度匹配
   可以在生成预测后添加验证代码，提前确认长度是否一致：

   # 训练集验证
   cat("训练集实际值长度：", length(train.data$target), "\n")
   cat("训练集预测值长度：", length(train.data.predict), "\n")
   # 测试集验证
   cat("测试集实际值长度：", length(test.data$target), "\n")
   cat("测试集预测值长度：", length(test.data.predict), "\n")
   

修复后的完整代码 #

set.seed(6522048)
library(randomForest)
# 注意：若因变量是heart_data，将此处的target替换为heart_data
model_rf1 <- randomForest(target ~ age+sex+cp+trestbps+chol+restecg+exang+ca, 
                          data = train.data, ntree = 2)
print("======================================================================================================================")
print('Confusion Matrix: TRAINING set based on random forest model built using 2 trees')
train.data.predict <- predict(model_rf1, train.data, type = "class")

conf.matrix <- table(train.data$target, train.data.predict)[,c('YES','NO')]
rownames(conf.matrix) <- paste("Actual", rownames(conf.matrix), sep = ": ")
colnames(conf.matrix) <- paste("Prediction", colnames(conf.matrix), sep = ": ")

# 打印格式化后的混淆矩阵
print(format(conf.matrix, justify="centre", digit=2))

print("=====================================================================================================================")
print('Confusion Matrix: TESTING set based on random forest model built using 2 trees')
test.data.predict <- predict(model_rf1, test.data, type = "class")

conf.matrix <- table(test.data$target, test.data.predict)[,c('YES','NO')]
rownames(conf.matrix) <- paste("Actual", rownames(conf.matrix), sep = ": ")
colnames(conf.matrix) <- paste("Prediction", colnames(conf.matrix), sep = ": ")

# 打印格式化后的混淆矩阵
print(format(conf.matrix, justify="centre", digit=2))

内容的提问来源于stack exchange，提问作者Jeff Routson