filter_all дает мне тот же нефильтрованный фрейм данных

Question

Я работаю с этим фреймом данных

structure(list(src_subject_id = c(19039L, 19040L, 19041L, 19042L, 
19043L, 19046L), interview_date = c("1/9/20", "9/4/19", "6/12/19", 
"6/17/19", "9/7/19", "11/27/19"), gender = c("F", "F", "F", "F", 
"F", "F"), interview_age = c("14", "13", "14", "13", "13", "13"
), MDDPast = c(1L, 1L, 1L, 1L, 1L, 1L), MDDCurrent = c(1L, 1L, 
1L, 1L, 1L, 1L), DysthymiaPast = c(1L, 1L, 1L, 1L, 1L, 1L), DysthymiaCurrent = c(1L, 
1L, 1L, 1L, 1L, 1L), DepNOSPast = c(1L, 1L, 1L, 1L, 1L, 1L), 
    DepNOSCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), CyclothymiaPast = c(1L, 
    1L, 1L, 1L, 1L, 1L), CyclothymiaCurrent = c(1L, 1L, 1L, 1L, 
    1L, 1L), BipolarNOSPast = c(1L, 1L, 1L, 1L, 1L, 1L), BipolarNOSCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), BipolarIPast = c(1L, 1L, 1L, 1L, 1L, 
    1L), BipolarICurrent = c(1L, 1L, 1L, 1L, 1L, 1L), BipolarIIPast = c(1L, 
    1L, 1L, 1L, 1L, 1L), BipolarIICurrent = c(1L, 1L, 1L, 1L, 
    1L, 1L), PsychoticFeaturesPast = c(1L, 1L, 1L, 1L, 1L, 1L
    ), PsychoticFeaturesCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), 
    PanicDisorderPast = c(1L, 1L, 1L, 1L, 1L, 1L), PanicDisorderCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), SeparationPast = c(1L, 1L, 1L, 1L, 1L, 
    1L), SeparationCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), ParentChildRelationalPast = c(1L, 
    1L, 1L, 1L, 1L, 1L), ParentChildRelationalCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), SimplePhobiaPast = c(1L, 1L, 1L, 1L, 
    1L, 1L), SimplePhobiaCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), 
    SocialPhobiaPast = c(1L, 1L, 1L, 1L, 1L, 1L), SocialPhobiaCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), AgoraphobiaPast = c(1L, 1L, 1L, 1L, 
    1L, 1L), AgoraphobiaCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), 
    AdjustmentDisorderPast = c(1L, 1L, 1L, 1L, 1L, 1L), AdjustmentDisorderCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), GADPast = c(1L, 1L, 3L, 1L, 1L, 1L), 
    GADCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), OCDPast = c(1L, 1L, 
    1L, 1L, 1L, 1L), OCDCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), 
    PTSDPast = c(1L, 1L, 1L, 1L, 1L, 1L), PTSDCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), AnxietyNOSPast = c(1L, 4L, 1L, 1L, 1L, 
    1L), AnxietyNOSCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), ADHDPast = c(1L, 
    1L, 1L, 1L, 1L, 1L), ADHDCurrent = c(1L, 1L, 1L, 1L, 3L, 
    1L), ODDPast = c(1L, 1L, 1L, 1L, 1L, 1L), ODDCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), ConductPast = c(1L, 1L, 3L, 1L, 1L, 
    1L), ConductCurrent = c(1L, 1L, 1L, 1L, 1L, 1L), DisruptiveBehPast = c(1L, 
    1L, 1L, 1L, 1L, 1L), DisruptiveBehCurrent = c(1L, 1L, 1L, 
    1L, 1L, 1L), AnorexiaPast = c(1L, 4L, 1L, 3L, 1L, 1L), AnorexiaCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), BulimiaPast = c(1L, 3L, 1L, 1L, 4L, 
    1L), BulimiaCurrent = c(1L, 1L, 1L, 1L, 1L, 4L), EatingNOSPast = c(1L, 
    1L, 1L, 1L, 1L, 1L), EatingNOSCurrent = c(1L, 1L, 1L, 1L, 
    1L, 1L), NicotinePast = c(1L, 1L, 1L, 3L, 1L, 1L), NicotineCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), AlcoholAbusePast = c(1L, 4L, 1L, 1L, 
    1L, 1L), AlcoholAbuseCurrent = c(1L, 1L, 1L, 3L, 1L, 1L), 
    AlcoholDependencePast = c(1L, 1L, 1L, 1L, 1L, 1L), AlcoholDependenceCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L), SubstanceAbusePast = c(1L, 3L, 1L, 1L, 
    1L, 1L), SubstanceAbuseCurrent = c(1L, 1L, 1L, 1L, 1L, 1L
    ), SubstanceDependencePast = c(1L, 1L, 1L, 4L, 1L, 3L), SubstanceDependenceCurrent = c(1L, 
    1L, 1L, 1L, 1L, 1L)), row.names = c(NA, 6L), class = "data.frame")

Назовите его fuKSADS_1. Я пытаюсь сделать fuKSADS_1_currentdx с помощью фильтрации, чтобы включить только значения 3 или 4. Я делаю это с помощью функции filter_all:

fuKSADS_1_currentdx <- filter_all(fuKSADS_1, vars(ends_with('Current')), .vars_predicate = any_vars(.>2))

Где я пытаюсь найти только текущие диагнозы и отфильтровать любые значения больше 2. Я думаю, что я не обязательно поступаю правильно, но это та строка кода, которая у меня есть прямо сейчас. С тем, что я написал, я только что вернул фрейм данных с именем fuKSADS_1_currentdx, который в точности совпадает с fuKSADS_1. Не совсем уверен, откуда у меня проблема.

Answer 1

В базовом R вы можете использовать rowSums в столбцах, заканчивающихся на "Current":

cols <- grep('Current$', names(fukSADS_1))
result <- fukSADS_1[rowSums(fukSADS_1[cols] > 2) > 0, cols]

Answer 2

Нам нужно filter_at вместо filter_all

library(dplyr)
fuKSADS_1%>% 
     select(ends_with('Current')) %>%
     filter_all(any_vars( . > 2))

Или другой вариант: c_across

fuKSADS_1 %>%
     select(ends_with('Current')) %>%
     rowwise %>% 
     filter(any(c_across(everything()) > 2)) %>%
     ungroup

В base R мы можем использовать Reduce

i1 <- endsWith(names(fuKSADS_1), 'Current')
subset(fuKSADS_1[i1], Reduce(`|`, lapply(fuKSADS_1[i1], `>`, 2)))