Question

我有一个带有int值的int列。我想将其转换为具有预定义数量的存储桶/级别/子范围的因子。这是一个示例：

dat1 <- fread('https://archive.ics.uci.edu/ml/machine-learning-databases/haberman/haberman.data',stringsAsFactors=T)
dat1 <- data.frame(lapply(dat1, as.factor))
> str (dat1)
'data.frame':   306 obs. of  4 variables:
 $ V1: Factor w/ 49 levels "30","31","33",..: 1 1 1 2 2 3 3 4 4 4 ...
 $ V2: Factor w/ 12 levels "58","59","60",..: 7 5 8 2 8 1 3 2 9 1 ...
 $ V3: Factor w/ 31 levels "0","1","2","3",..: 2 4 1 3 5 11 1 1 10 28 ...
 $ V4: Factor w/ 2 levels "1","2": 1 1 1 1 1 1 1 2 2 1 ...

我想将来源dat1$V3划分为多个范围（每个范围都是一个级别）。每个源值都属于这些类别之一。

Answer 1

使用-

dat1$V3_cut <- cut(as.numeric(dat1$V3), 5)

输出

  V1 V2 V3 V4   V3_cut
1 30 64  1  1 (0.97,7]
2 30 62  3  1 (0.97,7]
3 30 65  0  1 (0.97,7]
4 31 59  2  1 (0.97,7]
5 31 65  4  1 (0.97,7]
6 33 58 10  1   (7,13]

或

dat1$V3_cut <- cut(as.numeric(dat1$V3), c(0,3,5,11))

输出

  V1 V2 V3 V4 V3_cut
1 30 64  1  1  (0,3]
2 30 62  3  1  (3,5]
3 30 65  0  1  (0,3]
4 31 59  2  1  (0,3]
5 31 65  4  1  (3,5]
6 33 58 10  1 (5,11]

您可以指定cut的数量，也可以提供一个由类边界组成的列表。默认情况下，include.lowest是FALSE，由边界(]表示

编辑谢谢@Rui-

dat1$V3_cut <- cut(as.numeric(dat1$V3), c(0,3,5,11), labels=1:3)

限制从int到factor的转换级别

1 个答案: