我有一个df:
product store store1 review review1
book A B
shirt A B
pen A B
cd A B 0 2
dress A B 2 1
magazine A B 3 1
我希望store列中的值成为列名,我想在该列中插入审阅值,因此输出如下所示:
product A B
book 0 2
shirt 2 1
pen 3 1
此问题存在两个问题。首先,商店名称将来会发生很大变化,所以我不能使用这样的代码:
names(newdf)[names(newdf) == 'store'] <- 'a'
其次,我需要来自review和review1列的值从A和B列的第一行开始,所以说。例如,列a = book = 0,shirt = 2,magazine = 3.
我真的坚持这一点,任何帮助都会非常感激!
可重复的代码:
df <- data.frame(product = c("book","shirt", "pen", "cd", "dress", "magazine"), store=c("A", "A", "A", "A", "A", "A"),
store1=c("B", "B", "B", "B", "B", "B"), review=c("", "", "", 0, 2, 3), review1 =c("", "", "", 2, 1, 1))
答案 0 :(得分:1)
很难为此开发一种通用的方法,因为不清楚列的数量是否相同,或者它们是否可以一直一对一地匹配(例如,对于每个storeN
有reviewerN
)等。这是代码,你做了你想要的,但我不确定它是否符合目的。你应该更彻底地解释你的问题。
df <- data.frame(product = c("book","shirt", "pen", "cd", "dress", "magazine"), store=c("A", "A", "A", "A", "A", "A"),
store1=c("B", "B", "B", "B", "B", "B"), review=c("", "", "", 0, 2, 3), review1 =c("", "", "", 2, 1, 1))
# Convert factors to character
df <- data.frame(lapply(df, as.character), stringsAsFactors=FALSE)
# Blanks to NAs
df[df==""] <- NA
# Indicate which columns contain values to rename other columns
cols_with_values <- c(2,3)
# save first row in these columns
new_columns_names <- as.character(df[1,cols_with_values])
# Kill them!
df[,cols_with_values] <- NULL
# Rename columns
names(df) <- c(names(df[1]), new_columns_names)
# Show rows without NAs
df[complete.cases(df), ]
product A B
4 cd 0 2
5 dress 2 1
6 magazine 3 1
至于你的第二个问题,这是一项疯狂的任务,因为它会使你的数据非常非常混乱。我的意思是,你怎么知道book
= 3,而不是shirt
?