Down- and Up-Sampling Imbalanced Data

downSample will randomly sample a data set so that all classes have the same frequency as the minority class. upSample samples with replacement to make the class distributions equal

downSample(x, y, list = FALSE, yname = "Class")

Arguments

x	a matrix or data frame of predictor variables
y	a factor variable with the class memberships
list	should the function return list(x, y) or bind x and y together? If TRUE, the output will be coerced to a data frame.
yname	if list = FALSE, a label for the class column

Value

Either a data frame or a list with elements x and y.

Details

Simple random sampling is used to down-sample for the majority class(es). Note that the minority class data are left intact and that the samples will be re-ordered in the down-sampled version.

For up-sampling, all the original data are left intact and additional samples are added to the minority classes with replacement.

Examples

## A ridiculous example...
data(oil)
table(oilType)
#> oilType
#>  A  B  C  D  E  F  G 
#> 37 26  3  7 11 10  2 
downSample(fattyAcids, oilType)
#>    Palmitic Stearic Oleic Linoleic Linolenic Eicosanoic Eicosenoic Class
#> 1      11.1     5.0  32.9     49.8       0.3        0.4        0.1     A
#> 2      10.5     5.0  31.8     51.3       0.4        0.4        0.1     A
#> 3       6.8     4.3  26.4     60.5       1.3        0.1        0.1     B
#> 4       6.0     4.4  27.1     60.9       0.9        0.1        0.1     B
#> 5       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 6       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 7       9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 8      12.0     2.7  75.1      8.5       0.8        0.1        0.1     D
#> 9      11.9     3.8  25.7     52.7       5.8        0.1        0.1     E
#> 10      9.6     3.5  30.3     49.2       5.9        0.4        0.3     E
#> 11      4.5     1.7  64.9     18.6       8.3        0.1        0.1     F
#> 12      5.4     2.0  53.2     28.9       7.3        0.6        1.3     F
#> 13     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 14     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G

upSample(fattyAcids, oilType)
#>     Palmitic Stearic Oleic Linoleic Linolenic Eicosanoic Eicosenoic Class
#> 1        9.7     5.2  31.0     52.7       0.4        0.4        0.1     A
#> 2       11.1     5.0  32.9     49.8       0.3        0.4        0.1     A
#> 3       11.5     5.2  35.0     47.2       0.2        0.4        0.1     A
#> 4       10.0     4.8  30.4     53.5       0.3        0.4        0.1     A
#> 5       12.2     5.0  31.1     50.5       0.3        0.4        0.1     A
#> 6        9.8     4.2  43.0     39.2       2.4        0.4        0.5     A
#> 7       10.5     5.0  31.8     51.3       0.4        0.4        0.1     A
#> 8       10.5     5.0  31.8     51.3       0.4        0.4        0.1     A
#> 9       11.5     5.2  35.0     47.2       0.2        0.4        0.1     A
#> 10      10.0     4.8  30.4     53.5       0.3        0.4        0.1     A
#> 11      11.1     5.0  32.9     49.8       0.3        0.4        0.1     A
#> 12       9.3     4.4  43.3     39.2       2.3        0.4        0.5     A
#> 13      12.2     5.0  31.1     50.5       0.3        0.4        0.1     A
#> 14       9.7     5.6  35.2     47.8       0.5        0.4        0.2     A
#> 15      11.1     4.6  28.3     50.6       4.2        0.4        0.2     A
#> 16      11.0     4.7  29.8     49.6       3.4        0.4        0.2     A
#> 17      11.6     6.0  35.4     45.9       0.2        0.4        0.1     A
#> 18       9.8     5.3  31.7     51.3       0.8        0.4        0.2     A
#> 19      11.2     6.2  37.7     43.6       0.2        0.5        0.2     A
#> 20       7.6     4.7  29.1     56.1       0.7        0.4        0.2     A
#> 21       7.7     4.7  31.1     54.3       0.8        0.4        0.2     A
#> 22       9.9     5.2  37.4     44.2       2.1        0.4        0.3     A
#> 23      11.5     5.1  27.8     54.5       0.2        0.4        0.1     A
#> 24      11.3     5.8  35.2     46.7       0.2        0.4        0.1     A
#> 25      12.2     5.4  29.4     53.0       1.0        0.1        0.1     A
#> 26      11.0     5.3  35.0     45.2       1.3        1.3        0.7     A
#> 27      11.9     5.6  33.6     48.9       1.0        0.1        0.1     A
#> 28      11.4     5.8  34.5     48.3       1.0        0.1        0.1     A
#> 29      10.7     5.4  39.3     43.2       1.4        0.1        0.1     A
#> 30      11.2     6.2  35.8     44.1       0.7        1.1        0.1     A
#> 31      11.4     5.8  33.9     48.1       0.8        0.8        0.1     A
#> 32      11.5     6.2  39.7     42.6       0.8        0.1        0.1     A
#> 33      13.0     6.2  25.8     55.0       0.8        0.1        0.1     A
#> 34      13.0     6.7  29.6     50.8       0.5        0.1        0.1     A
#> 35      13.1     6.3  26.5     53.6       0.5        0.5        0.1     A
#> 36      11.6     6.5  38.4     42.8       0.5        0.7        0.1     A
#> 37      13.1     5.7  31.7     49.5       0.6        0.1        0.1     A
#> 38       6.1     4.1  24.0     64.3       0.1        0.3        0.1     B
#> 39       6.2     3.9  27.1     59.9       1.3        0.3        0.3     B
#> 40       6.0     4.9  22.8     64.4       0.3        0.3        0.2     B
#> 41       7.2     4.5  25.6     61.1       0.2        0.3        0.2     B
#> 42       6.2     4.1  26.8     61.4       0.1        0.3        0.2     B
#> 43       6.1     4.0  25.2     63.2       0.1        0.2        0.2     B
#> 44       6.1     4.1  26.7     61.0       0.6        0.3        0.2     B
#> 45       6.2     4.0  25.8     62.2       0.4        0.3        0.2     B
#> 46       6.0     3.8  29.6     57.7       1.2        0.3        0.3     B
#> 47       6.5     2.8  23.2     66.1       0.1        0.2        0.2     B
#> 48       7.0     3.4  23.2     64.9       0.1        0.2        0.2     B
#> 49       6.0     4.0  28.3     60.1       0.1        0.3        0.2     B
#> 50       6.1     4.1  25.1     63.5       0.5        0.1        0.1     B
#> 51       6.3     4.2  27.4     61.4       0.8        0.1        0.1     B
#> 52       6.2     4.2  27.1     61.8       0.8        0.1        0.1     B
#> 53       6.2     4.2  27.0     60.9       0.5        0.3        0.1     B
#> 54       6.2     4.0  28.3     59.7       0.9        0.1        0.1     B
#> 55       5.6     4.2  25.7     58.9       1.7        2.8        0.9     B
#> 56       6.4     3.9  26.0     63.7       0.5        0.1        0.1     B
#> 57       6.8     4.3  26.4     60.5       1.3        0.1        0.1     B
#> 58       6.0     4.4  27.1     60.9       0.9        0.1        0.1     B
#> 59       6.4     4.8  25.3     61.8       1.0        0.1        0.1     B
#> 60       5.9     4.5  24.1     61.7       0.9        0.6        0.6     B
#> 61       6.2     4.1  29.9     57.8       1.3        0.1        0.1     B
#> 62       6.6     4.7  24.5     62.8       0.3        0.4        0.1     B
#> 63       6.4     4.4  24.4     63.7       0.4        0.4        0.1     B
#> 64       6.1     4.1  25.1     63.5       0.5        0.1        0.1     B
#> 65       7.0     3.4  23.2     64.9       0.1        0.2        0.2     B
#> 66       6.2     4.2  27.1     61.8       0.8        0.1        0.1     B
#> 67       6.1     4.0  25.2     63.2       0.1        0.2        0.2     B
#> 68       7.2     4.5  25.6     61.1       0.2        0.3        0.2     B
#> 69       7.2     4.5  25.6     61.1       0.2        0.3        0.2     B
#> 70       6.5     2.8  23.2     66.1       0.1        0.2        0.2     B
#> 71       6.1     4.1  24.0     64.3       0.1        0.3        0.1     B
#> 72       6.5     2.8  23.2     66.1       0.1        0.2        0.2     B
#> 73       6.4     3.9  26.0     63.7       0.5        0.1        0.1     B
#> 74       6.5     2.8  23.2     66.1       0.1        0.2        0.2     B
#> 75       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 76      10.0     3.3  60.0     21.3       0.2        1.5        1.3     C
#> 77       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 78      10.0     3.3  60.0     21.3       0.2        1.5        1.3     C
#> 79       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 80      10.0     3.3  60.0     21.3       0.2        1.5        1.3     C
#> 81       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 82       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 83       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 84      10.0     3.3  60.0     21.3       0.2        1.5        1.3     C
#> 85       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 86       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 87       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 88       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 89       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 90       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 91       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 92       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 93       9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 94       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 95       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 96       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 97       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 98       9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 99      10.0     3.3  60.0     21.3       0.2        1.5        1.3     C
#> 100      9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 101      9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 102      9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 103      9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 104      9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 105      9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 106     10.0     3.3  60.0     21.3       0.2        1.5        1.3     C
#> 107      9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 108      9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 109      9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 110      9.7     3.4  59.3     20.5       0.1        1.5        1.2     C
#> 111      9.6     3.3  57.7     20.7       0.2        1.5        1.8     C
#> 112     14.9     2.6  68.2     12.8       0.6        0.4        0.3     D
#> 113      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 114     10.9     2.7  76.7      7.9       0.8        0.1        0.1     D
#> 115     10.5     2.8  75.8      8.0       0.7        0.1        0.1     D
#> 116     12.0     2.7  75.1      8.5       0.8        0.1        0.1     D
#> 117     11.7     2.9  74.6     10.1       0.6        0.1        0.1     D
#> 118     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 119     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 120      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 121     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 122     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 123     14.9     2.6  68.2     12.8       0.6        0.4        0.3     D
#> 124     12.0     2.7  75.1      8.5       0.8        0.1        0.1     D
#> 125      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 126     14.9     2.6  68.2     12.8       0.6        0.4        0.3     D
#> 127     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 128     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 129     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 130     11.4     3.0  73.0     10.6       0.7        0.1        0.1     D
#> 131     12.0     2.7  75.1      8.5       0.8        0.1        0.1     D
#> 132     11.7     2.9  74.6     10.1       0.6        0.1        0.1     D
#> 133     10.9     2.7  76.7      7.9       0.8        0.1        0.1     D
#> 134     14.9     2.6  68.2     12.8       0.6        0.4        0.3     D
#> 135     10.9     2.7  76.7      7.9       0.8        0.1        0.1     D
#> 136      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 137     11.7     2.9  74.6     10.1       0.6        0.1        0.1     D
#> 138      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 139     10.5     2.8  75.8      8.0       0.7        0.1        0.1     D
#> 140      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 141      9.3     2.8  65.0     17.0       3.9        0.5        0.7     D
#> 142     12.0     2.7  75.1      8.5       0.8        0.1        0.1     D
#> 143     10.9     2.7  76.7      7.9       0.8        0.1        0.1     D
#> 144     12.0     2.7  75.1      8.5       0.8        0.1        0.1     D
#> 145     14.9     2.6  68.2     12.8       0.6        0.4        0.3     D
#> 146     10.5     2.8  75.8      8.0       0.7        0.1        0.1     D
#> 147     14.9     2.6  68.2     12.8       0.6        0.4        0.3     D
#> 148     10.9     2.7  76.7      7.9       0.8        0.1        0.1     D
#> 149     10.9     3.6  26.0     52.6       5.5        0.4        0.2     E
#> 150      9.6     3.5  30.3     49.2       5.9        0.4        0.3     E
#> 151     10.5     4.2  25.5     52.0       7.8        0.1        0.1     E
#> 152     10.0     4.2  24.9     53.2       6.9        0.4        0.1     E
#> 153     10.4     4.2  25.9     50.8       7.5        0.4        0.4     E
#> 154     10.5     4.2  24.4     52.1       7.5        0.4        0.1     E
#> 155     10.5     4.3  24.6     53.1       7.6        0.1        0.1     E
#> 156     10.2     4.0  23.1     55.1       7.1        0.5        0.1     E
#> 157     10.9     3.8  27.2     49.5       6.4        0.7        0.9     E
#> 158     11.9     3.8  25.7     52.7       5.8        0.1        0.1     E
#> 159      9.7     3.9  25.1     54.2       5.9        0.1        0.1     E
#> 160     10.2     4.0  23.1     55.1       7.1        0.5        0.1     E
#> 161      9.6     3.5  30.3     49.2       5.9        0.4        0.3     E
#> 162     10.9     3.6  26.0     52.6       5.5        0.4        0.2     E
#> 163     10.5     4.3  24.6     53.1       7.6        0.1        0.1     E
#> 164     10.2     4.0  23.1     55.1       7.1        0.5        0.1     E
#> 165     10.0     4.2  24.9     53.2       6.9        0.4        0.1     E
#> 166     10.0     4.2  24.9     53.2       6.9        0.4        0.1     E
#> 167      9.6     3.5  30.3     49.2       5.9        0.4        0.3     E
#> 168     10.5     4.2  24.4     52.1       7.5        0.4        0.1     E
#> 169     10.5     4.2  24.4     52.1       7.5        0.4        0.1     E
#> 170      9.7     3.9  25.1     54.2       5.9        0.1        0.1     E
#> 171     10.9     3.6  26.0     52.6       5.5        0.4        0.2     E
#> 172     10.5     4.2  25.5     52.0       7.8        0.1        0.1     E
#> 173     10.4     4.2  25.9     50.8       7.5        0.4        0.4     E
#> 174      9.6     3.5  30.3     49.2       5.9        0.4        0.3     E
#> 175     10.5     4.3  24.6     53.1       7.6        0.1        0.1     E
#> 176     10.9     3.6  26.0     52.6       5.5        0.4        0.2     E
#> 177     10.5     4.2  25.5     52.0       7.8        0.1        0.1     E
#> 178     10.2     4.0  23.1     55.1       7.1        0.5        0.1     E
#> 179      9.7     3.9  25.1     54.2       5.9        0.1        0.1     E
#> 180     10.5     4.3  24.6     53.1       7.6        0.1        0.1     E
#> 181     10.4     4.2  25.9     50.8       7.5        0.4        0.4     E
#> 182     10.9     3.6  26.0     52.6       5.5        0.4        0.2     E
#> 183     10.2     4.0  23.1     55.1       7.1        0.5        0.1     E
#> 184     10.2     4.0  23.1     55.1       7.1        0.5        0.1     E
#> 185      9.7     3.9  25.1     54.2       5.9        0.1        0.1     E
#> 186      5.1     2.3  55.9     27.4       6.8        0.5        0.5     F
#> 187      4.8     1.8  62.6     20.0       9.5        0.1        1.4     F
#> 188      5.5     1.7  59.0     21.3       9.3        0.6        1.5     F
#> 189      5.1     1.9  59.2     22.3       9.3        0.7        1.6     F
#> 190      4.8     1.9  61.6     20.9       8.0        0.8        1.5     F
#> 191      5.4     2.0  53.2     28.9       7.3        0.6        1.3     F
#> 192      5.1     1.9  59.2     22.4       9.3        0.6        1.5     F
#> 193      4.5     1.7  64.9     18.6       8.3        0.1        0.1     F
#> 194      5.7     2.1  54.6     26.8       8.0        0.1        0.1     F
#> 195      6.2     2.2  52.2     29.0       8.0        0.1        0.1     F
#> 196      5.7     2.1  54.6     26.8       8.0        0.1        0.1     F
#> 197      5.1     2.3  55.9     27.4       6.8        0.5        0.5     F
#> 198      4.5     1.7  64.9     18.6       8.3        0.1        0.1     F
#> 199      5.1     1.9  59.2     22.3       9.3        0.7        1.6     F
#> 200      4.8     1.8  62.6     20.0       9.5        0.1        1.4     F
#> 201      5.1     1.9  59.2     22.3       9.3        0.7        1.6     F
#> 202      5.4     2.0  53.2     28.9       7.3        0.6        1.3     F
#> 203      6.2     2.2  52.2     29.0       8.0        0.1        0.1     F
#> 204      5.5     1.7  59.0     21.3       9.3        0.6        1.5     F
#> 205      5.1     1.9  59.2     22.3       9.3        0.7        1.6     F
#> 206      5.1     2.3  55.9     27.4       6.8        0.5        0.5     F
#> 207      5.4     2.0  53.2     28.9       7.3        0.6        1.3     F
#> 208      5.4     2.0  53.2     28.9       7.3        0.6        1.3     F
#> 209      5.1     1.9  59.2     22.3       9.3        0.7        1.6     F
#> 210      5.7     2.1  54.6     26.8       8.0        0.1        0.1     F
#> 211      5.1     1.9  59.2     22.3       9.3        0.7        1.6     F
#> 212      5.4     2.0  53.2     28.9       7.3        0.6        1.3     F
#> 213      5.7     2.1  54.6     26.8       8.0        0.1        0.1     F
#> 214      4.8     1.8  62.6     20.0       9.5        0.1        1.4     F
#> 215      4.8     1.9  61.6     20.9       8.0        0.8        1.5     F
#> 216      6.2     2.2  52.2     29.0       8.0        0.1        0.1     F
#> 217      5.1     2.3  55.9     27.4       6.8        0.5        0.5     F
#> 218      5.1     1.9  59.2     22.4       9.3        0.6        1.5     F
#> 219      5.1     2.3  55.9     27.4       6.8        0.5        0.5     F
#> 220      4.5     1.7  64.9     18.6       8.3        0.1        0.1     F
#> 221      4.5     1.7  64.9     18.6       8.3        0.1        0.1     F
#> 222      5.5     1.7  59.0     21.3       9.3        0.6        1.5     F
#> 223     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 224     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 225     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 226     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 227     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 228     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 229     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 230     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 231     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 232     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 233     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 234     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 235     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 236     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 237     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 238     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 239     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 240     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 241     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 242     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 243     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 244     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 245     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 246     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 247     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 248     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 249     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 250     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 251     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 252     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 253     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 254     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 255     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 256     10.0     2.3  36.9     47.1       2.2        0.5        0.5     G
#> 257     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 258     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G
#> 259     10.7     1.8  30.2     55.5       0.9        0.5        0.3     G