3

所以我在 R Studio 中运行下面的代码并得到这个错误：

UseMethod("tbl_vars") 中的错误：没有适用于 'tbl_vars' 的方法应用于“字符”类的对象

我不知道如何解决它，因为没有 tbl_vars 函数！有人可以帮忙吗？

for (i in 1:ceiling(nrow(reviews)/batch)) {
    row_start <- i*batch-batch+1

    row_end <- ifelse(i*batch < nrow(reviews), i*batch, nrow(reviews))

    print(paste("Processing row", row_start, "to row", row_end))

    reviews[row_start:row_end, ] %>%
        unnest_tokens(word, text) -> reviews_subset

    reviews_subset$row <- 1:nrow(reviews_subset)

    reviews_subset %>%
        anti_join(stopwords) %>%
        arrange(row) -> reviews_subset

    write_feather(reviews_subset, path = paste0("reviews", i, ".txt"))
}

ps: dplyr 已安装。还有其他已安装的软件包：pacman、feather、data.table、devtools、tidyr、tidytext、tokenizers、tibble

我正在使用它来处理 Yelp 数据集。

非常感谢你，卡梅姆

ps2：数据集示例（经过编辑和简化以适合此处）：

> dput(as.data.frame(review))

structure(list(user_id = 1:10, review_id = 11:20, business_id = 21:30, 
    stars = c(2L, 2L, 5L, 4L, 4L, 5L, 4L, 3L, 5L, 4L), text = c("Are you the type of person that requires being seen in an expensive, overly pretentious restaurant so that you can wear it as a status symbol?  Or maybe you're a gansta who dresses like CiLo Green and wants to show the hunny's (yes, a group of them out with one man) a night on the town!", 
    "Today was my first visit to the new luna, and I was disappointed-- both because I really liked the old cafe luna, and because the new luna came well recommended", 
    "Stayed here a few months ago and still remember the great service I received.", 
    "I came here for a business lunch from NYC and had a VERY appetizing meal. ", 
    "Incredible food with great flavor. ", 
    "OMG, y'all, try the Apple Pie Moonshine.  It. Is. Seriously. Good.  Smoooooooth.   The best rum that I've sampled so far: Zaya.", 
    "Caitlin is an amazing stylist.  She took time to hear what I had to say before jumping in", 
    "Oh yeah! After some difficulties in securing dinner, my dad and I found ourselves at one of the billion Primanti's locations for a quick feast", 
    "I've been going to this studio since the beginning of January", 
    "The best cannoli, hands down!!"
    )), .Names = c("user_id", "review_id", "business_id", "stars", 
"text"), row.names = c(NA, -10L), class = "data.frame")
4

3 回答 3

3

更改anti_join(stopwords)为anti_join(stop_words). stopwords可能不存在或不是您想要的

于 2018-05-24T14:22:34.197 回答
2

这

UseMethod("tbl_vars") 中的错误：没有适用于 'tbl_vars' 的方法...

消息不是由缺少tbl_vars函数引起的。当我错误地将向量传递给 dplyr 连接函数而不是另一个数据帧时，我遇到了完全相同的错误。这是一个简单的示例，说明如何使用 dplyr 0.7.5 在 R 3.5 中生成此错误：

library(dplyr)
# Create a dataframe of sales by person and bike color
salesNames = c('Sally', 'Jim', 'Chris', 'Chris', 'Jim',
               'Sally', 'Jim', 'Sally', 'Chris', 'Sally')

salesDates = c('2018-06-01', '2018-06-05', '2018-06-10', '2018-06-15',
               '2018-06-20', '2018-06-25', '2018-06-30', '2018-07-09',
               '2018-07-12', '2018-07-14')

salesColor = c('red', 'red', 'red', 'green', 'red',
               'blue', 'green', 'green', 'green', 'blue')

df_sales = data.frame(Salesperson = salesNames,
                      SalesDate = as.Date(salesDates),
                      BikeColor = salesColor,
                      stringsAsFactors = F)

# Create another dataframe to join to
modelColor = c('red', 'blue', 'green', 'yellow', 'orange', 'black')
modelPrice = c(279.95, 269.95, 264.95, 233.54, 255.27, 289.95)
modelCommission = modelPrice * 0.20

df_commissions = data.frame(ModelColor = modelColor,
                            ModelPrice = modelPrice,
                            Commission = modelCommission,
                            stringsAsFactors = F)

df_sales_comm = df_sales %>% left_join(df_commissions,
                                       by = c('BikeColor'= 'ModelColor'))

这工作正常。现在试试这个：

df_comms = df_commissions$ModelColor  # vector instead of dataframe

df_sales_comm2 = df_sales %>% left_join(df_comms,
                                        by = c('BikeColor'= 'ModelColor'))

并且您应该看到与您报告的完全相同的错误，因为 df_comms 不是日期框。您遇到的问题是停用词是向量而不是数据框（或小标题）。

于 2018-12-25T20:11:34.293 回答
2

有几种方法可以解决此错误。正如 Szczepaniak 指出的那样，根本原因是试图将字符向量传递到需要数据帧或 tibble 的操作中。

选项 1：将字符向量转换为数据帧（或 tibble），然后在 anti_join 中使用。转换示例：

        `stopwords <- tibble(joinColumn = stopwords)`

选项 2：更改操作以接受字符向量。在这种情况下，我们可以使用 filter 代替 anti_join ，如下所示：

         `reviews_subset <- reviews_subset %>%
              filter(!joinColumn %in% stopwords) %>%
              arrange(row) -> reviews_subset`
于 2019-02-13T18:49:54.137 回答