我有一个lexicon包含 650 个单词的数据框 ，我想通过从 中随机选择单词来为 5 个发言者创建一系列随机单词列表lexicon。我希望通过 24 个月的数据收集来完成此任务，每个月都会采集不同大小的词汇样本。指定月份和词汇量大小的基本数据框是df1：
df1 <- data.frame(months=rep(1:24, times=5, each=1),
vocab_size=(sample(c(0:25), 120, replace=TRUE)),
Speaker=rep(c("A", "B", "C", "D", "E"), times=1, each=24))
list1 <- split(df1, f=df1$Speaker)
lexicon看起来像这样：
lexicon <- data.frame(c("a", "about", "above", "ain't", "all", "am", "an", "and",
"animal", "ankle", "ant" ,"any", "apple","applesauce",
"asleep", "at", "ate", "aunt", "auntie", "aunty's",
"awake", "away", "baa", "baby" , "baby+doll", "bad" ,
"ball", "balloon", "banana", "basket", "bat", "bath",
"bathing", "bathtub", "be", "beach", "bead", "bean",
"because", "bed", "beddy", "bee", "been", "behind",
"being", "belt", "bench", "bib", "bicycle", "big"))
我一直在尝试使用以下代码生成我之后的输出：
vocab_data <- lapply(list1, FUN=function(element) {
all_vocab <- slice_sample(lexicon, n=element$vocab_size, replace=TRUE)
})
但我收到以下错误消息
Error in `slice_sample()`:
! `n` must be a constant.
Caused by error in `element$vocab_size`:
! $ operator is invalid for atomic vectors
有没有一种方法可以以这种方式从数据框中提取不同大小的样本，为每个说话者创建每个月的词汇列表？
使用sample在mapply. 请注意，我仅用lexicon作向量。
> set.seed(42)
> by(df1, ~Speaker, \(x)
+ mapply(sample, list(lexicon), x$vocab_size, MoreArgs=list(replace=TRUE))
+ )
Speaker: A
[[1]]
[1] "bicycle" "bead" "a" "baby+doll" "ankle" "beach" "aunt" "bicycle"
[9] "bench" "baby" "an" "beach" "baby+doll" "bead" "belt" "aunty's"
[[2]]
[1] "bad" "big" "bench" "above"
[[3]]
character(0)
[[4]]
[1] "beddy" "baby+doll" "ball" "beach" "bead" "bat" "being" "all"
[9] "aunty's" "bathtub" "balloon" "bed" "above" "bathing" "bee" "baby"
[17] "basket" "been" "asleep" "away" "and" "beach" "ain't" "away"
[[5]]
[1] "aunt" "being" "balloon" "all" "ain't" "bathtub" "big" "be" "baby"
[[6]]
[1] "baa" "bicycle" "bad"
[[7]]
[1] "big" "am" "am" "about" "above" "awake" "about" "bean" "ankle" "bed"
[11] "all" "bathing" "bicycle" "because" "beach" "being" "bee"
[[8]]
[1] "animal" "banana" "any" "aunty's" "animal" "been" "be" "banana" "at" "bead"
[11] "balloon" "belt" "all" "balloon" "about" "aunt" "baby" "bicycle" "aunt" "all"
[21] "belt" "bed" "bed" "awake" "beach"
[[9]]
[1] "beach" "because" "bee" "aunt" "ball" "apple" "auntie" "bath" "at" "banana"
[11] "ate" "behind" "animal" "awake" "been" "bib"
[[10]]
[1] "and" "banana" "bean" "a" "apple" "applesauce" "all" "animal"
[9] "at" "bad" "been" "bee" "bee" "bathing"
[[11]]
[1] "banana" "baby+doll" "bath" "ate" "belt" "applesauce" "bench" "am"
[9] "bench" "been" "bead" "at" "bicycle" "bicycle" "big" "bat"
[17] "bathtub" "basket" "bath" "am" "away" "bean" "bat"
[[12]]
[1] "am" "aunty's" "asleep" "bathtub" "bee" "baa"
[[13]]
[1] "any" "bad" "beddy"
[[14]]
[1] "a" "about" "baby" "bathtub" "baby+doll" "balloon" "been" "applesauce"
[9] "bathtub" "about" "bat" "bath" "ball" "ankle" "balloon" "bead"
[17] "ankle" "bicycle" "bicycle" "all" "be" "applesauce" "applesauce" "balloon"
[[15]]
[1] "bathing" "bead" "be" "bad"
[[16]]
[1] "bat" "being" "been" "bath" "ate" "ball" "basket" "beach" "aunty's" "ant"
[11] "ate" "animal" "ate"
[[17]]
[1] "apple" "bat" "bicycle" "a" "ate" "bathing" "balloon" "about" "bat" "and"
[11] "at" "above" "bead" "apple" "any" "a" "basket" "bat" "bead"
[[18]]
[1] "bench" "am" "at" "ankle" "an" "animal" "beddy" "baby+doll"
[9] "above" "aunt" "bench" "balloon" "asleep" "banana" "aunty's" "baby+doll"
[17] "at" "baby" "bathing" "bean" "been" "beddy" "away" "bad"
[25] "aunty's"
[[19]]
[1] "being" "bead" "been" "asleep" "awake" "bathtub" "baby+doll" "bean"
[9] "a" "applesauce" "bib" "bicycle" "ant" "balloon" "aunt" "bathtub"
[17] "bee"
[[20]]
[1] "beach" "aunty's" "asleep" "an" "bicycle" "ball" "basket" "all" "away" "aunty's"
[11] "big" "beddy" "ant" "balloon"
[[21]]
[1] "bean" "bee"
[[22]]
[1] "animal" "baa" "bad" "beddy" "ant" "auntie" "balloon" "awake"
[[23]]
[1] "beach" "bee" "away" "baby+doll" "bicycle" "basket" "awake" "behind"
[9] "awake" "am" "bench" "beddy" "ankle" "baby" "big" "an"
[17] "applesauce" "any" "ankle" "baby+doll" "bead" "about" "be" "bathtub"
[[24]]
[1] "ball" "belt" "bat"
----------------------------------------------------------------------------------
Speaker: B
[[1]]
[1] "ball" "at" "beach" "baby"
[[2]]
[1] "bean" "awake" "aunty's" "ant" "bathing" "applesauce" "balloon" "bathing"
[9] "banana" "beach" "being" "bath"
[[3]]
[1] "bean" "away" "behind" "an"
[[4]]
[1] "behind" "all" "bat" "bath" "aunt" "because" "apple" "basket" "about" "balloon"
[11] "above" "bathing" "bench" "banana" "aunty's" "been" "animal" "ankle" "an"
[[5]]
[1] "balloon"
[[6]]
[1] "aunt" "baby+doll" "behind" "ankle" "animal" "auntie" "asleep"
[[7]]
[1] "belt" "above"
[[8]]
[1] "away" "aunty's" "bee" "bathtub" "baa" "ate" "aunt" "auntie"
[9] "belt" "ate" "basket" "ankle" "bead" "being" "big" "ankle"
[17] "bib" "bathtub" "ain't" "baby+doll" "bad" "because" "bathing" "belt"
[25] "away"
[[9]]
character(0)
[[10]]
[1] "behind" "auntie" "am" "behind" "baby+doll" "basket" "bee" "because"
[9] "baby+doll"
[[11]]
[1] "about" "bean" "animal" "ant" "am" "bat" "all" "applesauce"
[9] "bee" "bench" "a" "beddy" "ankle" "bicycle" "any" "away"
[17] "a" "been" "ain't" "above" "bean" "all" "bee"
[[12]]
[1] "aunty's" "bathtub" "baa" "beach" "aunty's" "animal" "ball" "belt" "belt" "above"
[[13]]
[1] "bat" "aunt" "ankle" "bathing" "bed" "aunty's" "bead" "am" "banana" "belt"
[11] "ain't" "ball" "animal" "bed"
[[14]]
[1] "baa" "bib" "bad" "any" "ankle" "belt" "been" "baa"
[9] "ball" "bathtub" "at" "belt" "baby+doll" "balloon" "asleep" "bicycle"
[17] "bean" "baa" "auntie" "ankle" "apple"
[[15]]
[1] "bean" "a" "ant" "being" "bad" "ankle" "being" "asleep"
[9] "ate" "bench" "and" "beach" "am" "ankle" "ball" "aunty's"
[17] "baby+doll" "baby" "at" "behind" "at" "above" "bath" "awake"
[25] "because"
[[16]]
[1] "above" "ankle" "awake" "bee" "an" "auntie" "banana"
[[17]]
[1] "being" "bee" "belt"
[[18]]
[1] "baby" "banana" "basket"
[[19]]
[1] "been" "being" "be" "bee" "bathtub" "because" "awake" "been" "been" "bad"
[11] "bathing" "am" "baa" "bee" "beddy" "behind" "being" "aunt" "bench" "awake"
[21] "bathing"
[[20]]
[1] "ant" "bed" "being" "applesauce" "about" "bean" "being" "be"
[9] "balloon" "being" "at" "at" "being" "being" "and" "being"
[17] "bean"
[[21]]
[1] "bean" "and" "bat" "bead" "baa" "beddy" "ankle" "ant" "ball" "baa"
[11] "bathing" "away"
[[22]]
[1] "asleep" "ain't" "because" "been"
[[23]]
[1] "ankle" "and" "bat"
[[24]]
[1] "balloon"
----------------------------------------------------------------------------------
Speaker: C
[[1]]
[1] "be" "being" "aunty's" "bib" "and" "bath" "above" "beddy" "bib" "ant"
[11] "apple" "because" "aunt" "because" "awake" "all" "bat" "aunty's" "aunt" "because"
[21] "bicycle" "above" "bib"
[[2]]
[1] "and" "auntie" "big" "ankle" "asleep" "any" "be" "bib" "bath" "bee"
[11] "bathtub" "bed" "a" "apple" "an" "ant" "ant"
[[3]]
[1] "apple" "bead"
[[4]]
[1] "bathtub" "asleep" "bead" "applesauce" "bead" "bad" "bicycle" "balloon"
[9] "aunt" "ate" "big" "because" "being" "bathtub" "any" "basket"
[17] "be" "aunty's" "bad" "baby+doll" "and" "aunty's" "balloon"
[[5]]
[1] "belt" "bathing" "belt" "awake" "big" "bathtub" "all" "bee" "any" "above"
[11] "aunt" "big" "big" "bench" "a" "and" "bed" "basket" "aunty's" "bathtub"
[21] "awake" "auntie"
[[6]]
[1] "been" "bench" "aunt" "bed" "baby" "bee" "banana" "bathtub" "away" "ain't"
[11] "baa" "ate" "behind" "aunty's" "because" "animal"
[[7]]
[1] "auntie" "awake" "been" "been" "behind" "ate" "bathtub" "an" "bicycle" "ain't"
[11] "being" "a" "balloon" "ate" "beach" "bathtub" "ball" "banana" "bad" "awake"
[[8]]
[1] "bed" "be" "apple" "bathtub" "beach" "beddy" "been" "balloon"
[9] "ankle" "bath" "belt" "ball" "bathing" "a" "applesauce" "bath"
[17] "balloon" "been" "bath" "beddy" "basket" "bead" "bathtub" "beddy"
[25] "be"
[[9]]
[1] "baby" "behind" "bathtub" "baby" "balloon" "bib" "aunt" "basket" "away" "bad"
[11] "bicycle" "aunty's" "bat" "bed" "basket" "and" "bench"
[[10]]
[1] "baby+doll" "bean" "be" "a" "basket"
[[11]]
[1] "bench" "ain't" "balloon" "beddy" "bathing"
[[12]]
[1] "been"
[[13]]
[1] "am" "aunty's" "applesauce" "animal" "balloon" "bat" "baby" "am"
[9] "behind" "away" "bicycle" "ant" "animal" "aunt" "been" "auntie"
[17] "bathing" "been" "bicycle"
[[14]]
[1] "bicycle" "bed"
[[15]]
[1] "ant" "big" "bathing" "bean" "bathtub" "belt" "bead" "a"
[9] "basket" "aunt" "ain't" "asleep" "an" "applesauce" "ankle" "be"
[17] "an" "all" "baby" "bean" "all"
[[16]]
[1] "awake" "big" "beddy" "be" "above" "animal" "being" "ball" "banana" "ball"
[11] "because" "bee" "ate" "bathtub" "aunty's" "bicycle" "bad" "any" "baby" "baby"
[[17]]
[1] "bean"
[[18]]
[1] "basket" "away" "basket" "ankle" "bib" "bathing" "bee" "an"
[9] "baby+doll" "baby" "about" "beach" "and" "bench" "applesauce" "ate"
[17] "bead" "bicycle" "ate" "at" "bath" "baby+doll"
[[19]]
[1] "bean" "above" "applesauce" "animal" "behind"
[[20]]
[1] "animal" "baby" "belt" "above" "beach" "be" "because" "bathing" "bead" "away"
[11] "ain't" "banana" "ball" "and" "behind" "behind" "any" "bench" "animal" "baby"
[21] "above" "above" "bad" "beddy" "an"
[[21]]
[1] "ball" "baa" "behind" "being" "bicycle" "bath" "asleep" "bench" "aunty's"
[[22]]
[1] "be" "ball" "being" "auntie" "about" "away" "bead"
[[23]]
[1] "beddy" "big" "bathing" "ain't"
[[24]]
character(0)
----------------------------------------------------------------------------------
Speaker: D
[[1]]
[1] "aunt" "bead" "be" "bee" "baby+doll" "because" "applesauce" "ball"
[9] "ankle" "baa" "ankle" "awake" "beddy" "because" "am" "balloon"
[[2]]
[1] "and" "bat" "bad" "baby" "bead" "auntie"
[[3]]
[1] "aunty's" "ant" "ball"
[[4]]
[1] "bicycle" "an" "bee" "behind" "ain't" "bat" "be" "a" "bib" "all"
[11] "ate" "been"
[[5]]
[1] "a" "baby" "bathing" "aunt" "above" "all" "because" "am" "at"
[[6]]
[1] "apple" "bathing" "big" "ankle" "applesauce" "bee" "bathtub" "at"
[[7]]
[1] "beddy" "ate" "about" "asleep" "baa" "ball" "bed" "above" "bathtub" "bicycle"
[11] "above"
[[8]]
[1] "been" "bead" "away" "bathing" "above" "applesauce" "bean" "at"
[9] "ankle" "ball" "and" "baa" "animal" "be" "bathtub" "bean"
[17] "bathtub" "bib" "basket"
[[9]]
[1] "baby+doll" "basket" "animal" "bead" "ate" "behind" "aunty's" "beach"
[[10]]
[1] "bean" "bathtub" "above" "big" "ate" "baby" "big" "ant" "and" "because"
[[11]]
[1] "aunty's" "balloon"
[[12]]
[1] "baa" "bathing" "auntie" "at" "bed" "big" "ankle" "ankle" "auntie" "apple"
[11] "baa" "behind" "bed" "bean" "because"
[[13]]
[1] "because" "bead" "a" "been"
[[14]]
[1] "bathtub" "big" "baa" "bee" "bean" "beach" "any" "am"
[9] "ate" "belt" "a" "been" "aunty's" "bad" "auntie" "bead"
[17] "above" "being" "auntie" "big" "aunty's" "bat" "applesauce"
[[15]]
[1] "bathtub" "bathing" "away" "belt" "an" "bench" "bean" "behind"
[9] "been" "ankle" "basket" "bicycle" "baby+doll"
[[16]]
[1] "ankle" "basket" "ball" "auntie"
[[17]]
[1] "bee" "animal" "away" "ball" "and" "auntie" "an" "bad" "at" "and"
[11] "baa" "awake" "at" "bat" "banana" "been" "ball" "ankle" "aunty's" "aunt"
[21] "at"
[[18]]
[1] "and"
[[19]]
[1] "awake" "bead" "baa" "big" "bench" "about" "bathing" "balloon" "bat" "balloon"
[11] "bee" "animal" "bathing" "ate" "being" "am" "any"
[[20]]
[1] "auntie" "bathing" "ant" "balloon" "ball" "at" "applesauce" "baa"
[9] "bead" "bed" "auntie" "bib" "behind" "bench" "an" "bathing"
[17] "baby" "asleep" "and" "am" "been" "bathtub" "bathing"
[[21]]
[1] "basket" "bead" "bead" "been" "ball" "ball" "bicycle" "ate" "ate" "bad"
[11] "basket" "ball" "auntie" "apple" "aunt" "ankle"
[[22]]
[1] "above" "bicycle" "balloon" "awake" "ankle" "bathing" "ant" "apple"
[9] "belt" "being" "an" "bed" "beach" "applesauce" "all" "aunty's"
[17] "baa"
[[23]]
[1] "belt" "auntie" "bench" "bib"
[[24]]
[1] "above" "about" "bean" "because" "ankle" "being" "bee" "bee" "balloon" "be"
[11] "baby" "bicycle" "big"
----------------------------------------------------------------------------------
Speaker: E
[[1]]
[1] "ate" "ankle" "bee" "bed" "baby+doll" "baa" "apple"
[[2]]
[1] "bed" "ball" "bead" "ate" "baby+doll" "animal" "bed" "balloon"
[9] "auntie" "apple" "bathing" "any" "a" "bean" "auntie" "being"
[17] "auntie" "bead" "baa" "ball" "bat" "away"
[[3]]
[1] "apple" "applesauce" "belt" "behind" "aunty's" "and" "been"
[[4]]
[1] "applesauce" "about" "applesauce" "basket" "ankle" "about" "bee" "bath"
[9] "above" "bathtub" "balloon" "balloon" "basket" "ankle" "baby+doll" "bench"
[17] "auntie" "any" "aunt" "bat"
[[5]]
[1] "aunt" "been" "animal"
[[6]]
[1] "and" "beddy" "and" "be" "and" "ankle" "ankle" "bathtub"
[9] "ant" "bee" "bib" "aunty's" "about" "bat" "baby+doll" "ankle"
[17] "belt" "bath" "beach" "bib" "asleep" "baa" "baby" "baby+doll"
[[7]]
[1] "ant" "auntie" "bee"
[[8]]
[1] "bathtub" "all" "auntie" "baa" "been" "big"
[[9]]
[1] "been" "bee" "all" "being" "awake" "bee" "ate" "be" "belt"
[[10]]
[1] "ankle" "balloon" "bathtub" "away" "all" "baby" "any" "asleep"
[9] "bad" "bean" "bathing" "balloon" "baa" "being" "beddy" "applesauce"
[17] "any"
[[11]]
[1] "aunt" "an" "at" "any" "behind" "am" "auntie" "big" "about" "and" "baby"
[12] "awake"
[[12]]
[1] "bathing" "baby+doll" "bench" "balloon" "big" "belt" "baby+doll" "bathing"
[9] "am" "auntie" "beach" "awake" "a" "a" "a" "asleep"
[17] "bad" "ball" "aunt" "awake"
[[13]]
[1] "basket" "ant" "bathing" "basket" "ankle" "beddy" "bead" "bean" "been" "bead"
[11] "auntie" "am" "and" "at" "ate" "behind" "bee" "and" "bed" "because"
[21] "belt"
[[14]]
[1] "bench" "bench" "ant" "bee" "big" "asleep" "bat" "bench"
[9] "auntie" "being" "baa" "baby+doll" "above" "balloon" "bee" "at"
[17] "awake" "being"
[[15]]
[1] "ball" "bat" "big" "bathing" "animal" "been" "bead" "belt" "ball" "be"
[11] "all" "balloon" "ant" "belt" "bib"
[[16]]
[1] "bead" "and" "bean" "awake" "ain't" "all" "baby" "any" "big" "because"
[11] "balloon" "being" "bee" "bib" "beach" "about" "any" "being" "a" "bench"
[21] "any"
[[17]]
[1] "bench" "bib" "bat" "bicycle" "bicycle" "a" "bicycle" "ankle"
[9] "bat" "above" "ate" "ain't" "baby+doll" "ant" "because" "ankle"
[[18]]
[1] "at" "an" "be" "bat" "big" "awake" "bad" "banana" "been" "bead" "bib"
[[19]]
[1] "apple" "ain't" "bench" "and" "be" "animal" "apple" "all"
[[20]]
[1] "any" "ain't" "baby+doll" "ankle" "above" "because" "bean" "ate"
[9] "bad" "aunty's" "bee" "aunty's" "any" "aunty's" "balloon" "be"
[17] "big" "above" "bench" "any" "bench" "be" "bat" "aunty's"
[[21]]
[1] "baa" "bean" "ankle" "bad" "bat" "bean" "been" "a"
[9] "bee" "be" "apple" "all" "bee" "bib" "applesauce" "ball"
[17] "ball" "bathing" "bath" "a"
[[22]]
[1] "bath" "been" "applesauce" "bathing" "aunty's" "above" "above" "bicycle"
[9] "baa" "belt"
[[23]]
[1] "bathing" "am" "am" "applesauce" "bicycle" "aunty's" "and" "animal"
[9] "beach" "because" "beach" "asleep" "an" "banana" "asleep" "basket"
[17] "at" "being" "ate" "bench" "a" "auntie" "and" "bib"
[[24]]
[1] "belt" "ant" "ankle" "bench" "ball" "balloon" "big" "about" "bad" "and"
[11] "basket" "balloon" "at" "bath" "baa" "bead" "belt" "an" "bead" "any"
[21] "am" "basket" "ate" "bib" "baa"
数据：
set.seed(42)
df1 <- data.frame(months=rep(1:24, times=5, each=1),
vocab_size=(sample(c(0:25), 120, replace=TRUE)),
Speaker=rep(c("A", "B", "C", "D", "E"), times=1, each=24))
> dput(lexicon)
c("a", "about", "above", "ain't", "all", "am", "an", "and", "animal",
"ankle", "ant", "any", "appl
您可以使用该dplyr包创建一个大数据框架，其中包含每个说话者和月份的单词，如下所示：
library(dplyr)\n word_list <- df1 %>%\n group_by(months, Speaker) %>%\n reframe(words = sample(lexicon[[1]], size = vocab_size, replace = TRUE))\n然后，您可以获得给定说话者和月份的单词，如下所示：
\nfilter(word_list, Speaker == "A", months == 17)\n## # A tibble: 8 \xc3\x97 3\n## months Speaker words \n## <int> <chr> <chr> \n## 1 17 A baby+doll\n## 2 17 A any \n## 3 17 A been \n## 4 17 A aunt \n## 5 17 A baby \n## 6 17 A a \n## 7 17 A and \n## 8 17 A am \n如果您喜欢数据框列表，您可以轻松地从数据框创建它：
\nword_list2 <- split(word_list, interaction(word_list$Speaker, word_list$months))\nword_list2[["A.17"]]\n## # A tibble: 8 \xc3\x97 3\n## months Speaker words \n## <int> <chr> <chr> \n## 1 17 A baby+doll\n## 2 17 A any \n## 3 17 A been \n## 4 17 A aunt \n## 5 17 A baby \n## 6 17 A a \n## 7 17 A and \n## 8 17 A am \n如果你只想要单词，你可以按如下方式获取：
\nword_list3 <- lapply(word_list2, getElement, "words")\nword_list3[["A.17"]]\n## [1] "baby+doll" "any" "been" "aunt" "baby" "a" "and" "am" \n发生错误是因为在这种情况下slice_sample得到一个数字向量，但它只需要一个数字n。通过使用 来解决这个问题rowwise，它分别作用于每一行，因此slice_sample只能看到一个数字。
library(dplyr)\n\ndf1 %>% \n rowwise() %>% \n mutate(all_vocab = list(slice_sample(lexicon, n=vocab_size))) %>% \n ungroup()\n# A tibble: 120 \xc3\x97 4\n months vocab_size Speaker all_vocab \n <int> <int> <chr> <list> \n 1 1 1 A <df [1 \xc3\x97 1]> \n 2 2 15 A <df [15 \xc3\x97 1]>\n 3 3 18 A <df [18 \xc3\x97 1]>\n 4 4 18 A <df [18 \xc3\x97 1]>\n 5 5 24 A <df [24 \xc3\x97 1]>\n 6 6 4 A <df [4 \xc3\x97 1]> \n 7 7 3 A <df [3 \xc3\x97 1]> \n 8 8 10 A <df [10 \xc3\x97 1]>\n 9 9 24 A <df [24 \xc3\x97 1]>\n10 10 19 A <df [19 \xc3\x97 1]>\n# \xe2\x84\xb9 110 more rows\n# \xe2\x84\xb9 Use `print(n = ...)` to see more rows\n