我有一个DataFrame,它完全由数字数据类型的变量组成。我过去有一个很好的例程,可以检查DataFrame中每个变量的统计异常值,并用NA值替换所有已识别的异常值。但是,此例程利用了最近不建议使用的funs()。
研究了这个问题之后,我知道您应该可以用例如list(〜example_func())代替funs():
>funs(mean(., trim = .2), median(., na.rm = TRUE))
>
>Would become:
>
>list(~ mean(., trim = .2), ~ median(., na.rm = TRUE))
不幸的是,这种补救措施在我的用例中不起作用。
以下代码可以正常工作,如下所示(对于具有异常值的变量,将异常值替换为NA值);但是,它会触发有关现在已弃用的funs()的警告:
> # Which variables have missing values
> sapply(training_imptd, function(x) sum(is.na(x)))
INDEX TARGET_WINS TEAM_BATTING_H TEAM_BATTING_2B TEAM_BATTING_3B
0 0 0 0 0
TEAM_BATTING_HR TEAM_BATTING_BB TEAM_BATTING_SO TEAM_BASERUN_SB TEAM_BASERUN_CS
0 0 102 131 772
TEAM_BATTING_HBP TEAM_PITCHING_H TEAM_PITCHING_HR TEAM_PITCHING_BB TEAM_PITCHING_SO
2085 0 0 0 102
TEAM_FIELDING_E TEAM_FIELDING_DP
0 286
>
> # Identify outliers and set them to NA (NAs to be fixed in next step by mice)
> training_imptd <- training_imptd %>%
+ mutate_all(
+ funs(ifelse(. %in% boxplot.stats(training_imptd$.)$out, NA, .))
+ )
>
> Warning: funs() is soft deprecated as of dplyr 0.8.0
> Please use a list of either functions or lambdas:
>
> # Simple named list:
> list(mean = mean, median = median)
>
> # Auto named with `tibble::lst()`:
> tibble::lst(mean, median)
>
> # Using lambdas
> list(~ mean(., trim = .2), ~ median(., na.rm = TRUE))
> This warning is displayed once per session.
>
> # Which variables have missing values (after imputing NA for outliers)
> sapply(training_imptd, function(x) sum(is.na(x)))
INDEX TARGET_WINS TEAM_BATTING_H TEAM_BATTING_2B TEAM_BATTING_3B
0 32 67 15 29
TEAM_BATTING_HR TEAM_BATTING_BB TEAM_BATTING_SO TEAM_BASERUN_SB TEAM_BASERUN_CS
0 129 102 252 827
TEAM_BATTING_HBP TEAM_PITCHING_H TEAM_PITCHING_HR TEAM_PITCHING_BB TEAM_PITCHING_SO
2086 213 4 90 140
TEAM_FIELDING_E TEAM_FIELDING_DP
303 318
根据我所读到的有关将funs()替换为list(〜example_func())的信息,我希望以下代码与上面利用funs()的代码完全一样,但不会(对于带有离群值的变量,离群值不会被NA值取代):
> # Which variables have missing values
> sapply(training_imptd, function(x) sum(is.na(x)))
INDEX TARGET_WINS TEAM_BATTING_H TEAM_BATTING_2B TEAM_BATTING_3B
0 0 0 0 0
TEAM_BATTING_HR TEAM_BATTING_BB TEAM_BATTING_SO TEAM_BASERUN_SB TEAM_BASERUN_CS
0 0 102 131 772
TEAM_BATTING_HBP TEAM_PITCHING_H TEAM_PITCHING_HR TEAM_PITCHING_BB TEAM_PITCHING_SO
2085 0 0 0 102
TEAM_FIELDING_E TEAM_FIELDING_DP
0 286
>
> # Identify outliers and set them to NA (NAs to be fixed in next step by mice)
> training_imptd <- training_imptd %>%
+ mutate_all(
+ list(~ ifelse(. %in% boxplot.stats(training_imptd$.)$out, NA, .))
+ )
>
> # Which variables have missing values (after imputing NA for outliers)
> sapply(training_imptd, function(x) sum(is.na(x)))
INDEX TARGET_WINS TEAM_BATTING_H TEAM_BATTING_2B TEAM_BATTING_3B
0 0 0 0 0
TEAM_BATTING_HR TEAM_BATTING_BB TEAM_BATTING_SO TEAM_BASERUN_SB TEAM_BASERUN_CS
0 0 102 131 772
TEAM_BATTING_HBP TEAM_PITCHING_H TEAM_PITCHING_HR TEAM_PITCHING_BB TEAM_PITCHING_SO
2085 0 0 0 102
TEAM_FIELDING_E TEAM_FIELDING_DP
0 286
答案 0 :(得分:1)
从函数内部删除不必要的training_imptd$
。代词.
已经指向“当前列”,因此您可以将其直接传递给boxplot.stats()
:
training_imptd %>%
mutate_all(
~ifelse(. %in% boxplot.stats(.)$out, NA, .)
)