条件校验-稀疏数据

Question

我想检查一个条件，如果==为== 'value‘，那么如果从列列表中，则为任何列== 'value2’。

# create dummy data set
pb=c('1','0','0','0','0','1','Not_ans','1','0','Not_ans')
qa=c('1','1','0','0','1','0','Not_ans','1','Not_ans','Not_ans')
#zy=c('1','Not_ans','0','1','Not_ans','0','1','1','1','Not_ans')

#sub questions for pb
pb.abr=c('1','0','0','0','0','1','0','1','0','0')
pb.ras=c('0','0','0','0','1','0','0','1','0','0')
pb.sfg=c('1','0','0','0','0','0','0','1','0','0')

#sub questions for qa
qa.fgs=c('1','0','0','0','0','0','0','1','0','0')
qa.sdf=c('0','1','0','0','0','0','0','0','0','0')
qa.tyu=c('0','0','0','0','1','0','0','1','0','0')

df=data.frame(pb,qa,pb.abr,pb.ras,pb.sfg,qa.fgs,qa.sdf,qa.tyu)
df

        pb      qa     pb.abr pb.ras pb.sfg qa.fgs qa.sdf qa.tyu
1        1       1      1      0      1      1      0      0
2        0       1      0      0      0      0      1      0
3        0       0      0      0      0      0      0      0
4        0       0      0      0      0      0      0      0
5        0       1      0      1      0      0      0      1
6        1       0      1      0      0      0      0      0
7  Not_ans Not_ans      0      0      0      0      0      0
8        1       1      1      1      1      1      0      1
9        0 Not_ans      0      0      0      0      0      0
10 Not_ans Not_ans      0      0      0      0      0      0

在上面的数据集中-我要检查的是如果列'pb‘是0，那么如果任何列pb.abr或pb.ras或pb.sfg == 1

subset_df=subset(df,(pb==0) & ((pb.abr==1) | (pb.ras==1)|(pb.sfg==1)))

挑战是，我有1000多个列，什么是pb.xxx格式的，手动编写子集中的所有列并不是一个可行的解决方案。如何使用使用contains("pb.")的逻辑检查上述内容，以及如何检查列之间具有OR条件的列，并最终提供数据框架？

Answer 1

我们可以用filter_at

library(dplyr)

df %>%
  filter(pb == 0) %>%
  filter_at(vars(matches("pb\\.")), any_vars(.  == 1))

#  pb qa pb.abr pb.ras pb.sfg qa.fgs qa.sdf qa.tyu
#1  0  1      0      1      0      0      0      1

或在基R中

df[df$pb == 0 & rowSums(df[grep("pb\\.", names(df))] == 1) > 0, ]

#  pb qa pb.abr pb.ras pb.sfg qa.fgs qa.sdf qa.tyu
#5  0  1      0      1      0      0      0      1

Answer 2

样本数据

dont' forget to set StringsAsFactors to FALSE!
df=data.frame(pb,qa,pb.abr,pb.ras,pb.sfg,qa.fgs,qa.sdf,qa.tyu, stringsAsFactors = FALSE)

码

library(dplyr)
df %>%
  #set all columns starting with 'pb.' to numeric
  mutate_at( vars( starts_with("pb.") ), funs( as.numeric ) ) %>%
  #first filter
  filter( pb == "0" ) %>%
  #second filter
  filter( rowSums( .[, grep("pb\\.", names(df))]) > 0 ) 

输出

pb pb.abr pb.ras pb.sfg qa.fgs qa.sdf qa.tyu 1 0 1 0 0 0 1

Answer 3

作为一条线：

filter(df,pb==0 & rowSums(z[,grepl("pb\\.",names(z))])>0)