根據R中另一個df2的信息添加列到dataframe

Question

假設我有 2 個這樣的數據框：

Model <- c("H5", "H5", "H5","H4","H3")
Code <- c("001001", "001002","001003","001004","001005")
City <-  c("Mexico", "London", "NY", "Otawa", "Liverpool")

df1 <- data.frame(Model,Length,Code)


Model   Code       City
H5      001001     Mexico  
H5      001002     London
H5      001003     NY
H4      001004     Otawa
H3      001005     Liverpool

和

X <- c("030299", "010121","030448","030324","010245","001001", "001002","001003","001004","001005")
Y <- c("030344", "010222","030448","030001","010245","221001", "221044","221044","221004"," 001005")
Var1 <- c("H5", "H5", "H4","H4","H4","H5", "H5", "H5","H4","H3")
Var2 <- c("H4", "H2", "H4","H3","H4","H3", "H3", "H3","H3","H3")

  df2 <- data.frame(X,Y,Var1,Var2)

  X            Y     VAR1   VAR2
030299      030344    H5     H4
010121      010222    H5     H2
030448      030448    H4     H4
030324      030001    H4     H3
010245      010245    H4     H4
001001      221001    H5     H3
001002      221044    H5     H3
001003      221044    H5     H3
001004      221004    H4     H3
001005      001005    H3     H3

我想編寫以下代碼：

例如，如果我 select H3 作為 function 中的參數，我想從 df1 的“代碼”列中獲取所有值，考慮到它在“模型”列中的相應值，並根據 df2 信息從“代碼”列轉換這些值. 例如，如果我們 select 來自 df1 的第一行並將 H3 設置為參數：

  H5      001001    Mexico 

function 必須從 df2 中獲取相應的行：

   X            Y     VAR1   VAR2
 001001      221001    H5     H3

然后像這樣給我 output：

   X            Y    VAR2  City   
 001001      221001   H3   Mexico   

最后的output應該是這樣的：

  X            Y     VAR2   City 

001001      221001    H3   Mexico  
001002      221044    H3   London  
001003      221044    H3   NY
001004      221004    H3   Otawa  
001005      221056    H3   Liverpool 

Answer 1

也許從一開始，這將重現您示例的結果。

df2 %>% 
  left_join(df1, by = c( "Var1" = "Model", "X" = "Code")) %>% 
  filter(Var2 == "H3", !is.na(City)) %>% 
  select(-Var1)

       X       Y Var2      City
1 001001  221001   H3    Mexico
2 001002  221044   H3    London
3 001003  221044   H3        NY
4 001004  221004   H3     Otawa
5 001005  001005   H3 Liverpool

Answer 2

按照您的邏輯，我嘗試創建一個基數為 R 的自定義 function：需要 3 個 arguments： df1, df2, x x是您要計算的行數。 因此，您可以 select 所有列或僅一列，如您在示例中所解釋的那樣。

my_function <- function(df1, df2, x){
select_row <- df1[x,]
cbind(df2[X==select_row[,2],c(1:2, 4)],select_row[3])
}

my_function(df1, df2, 1:5)

        X       Y Var2      City
6  001001  221001   H3    Mexico
7  001002  221044   H3    London
8  001003  221044   H3        NY
9  001004  221004   H3     Otawa
10 001005  001005   H3 Liverpool

Answer 3

像這樣？

library(data.table)
setDT(df1);setDT(df2)
df2[df1, on = .(Var1 = Model, X = Code)]
#         X       Y Var1 Var2      City
# 1: 001001  221001   H5   H3    Mexico
# 2: 001002  221044   H5   H3    London
# 3: 001003  221044   H5   H3        NY
# 4: 001004  221004   H4   H3     Otawa
# 5: 001005  001005   H3   H3 Liverpool

Answer 4

另一種方法：

數據

library(tidyverse)

Model <- c("H5", "H5", "H5", "H4", "H3")
Code <- c("001001", "001002", "001003", "001004", "001005")
City <- c("Mexico", "London", "NY", "Otawa", "Liverpool")

df1 <- data.frame(Model, Code, City)

X <- c("030299", "010121", "030448", "030324", "010245", "001001", "001002", "001003", "001004", "001005")
Y <- c("030344", "010222", "030448", "030001", "010245", "221001", "221044", "221044", "221004", " 001005")
Var1 <- c("H5", "H5", "H4", "H4", "H4", "H5", "H5", "H5", "H4", "H3")
Var2 <- c("H4", "H2", "H4", "H3", "H4", "H3", "H3", "H3", "H3", "H3")

df2 <- data.frame(X, Y, Var1, Var2)

Function

my_fun <- function(row, var2) {
  df1_data <- df1 %>% slice(row)
  df2 %>%
    filter(Var2 == var2 & X == df1_data$Code) %>%
    mutate(df1_data$City)
}

1:nrow(df1) %>%
  map_dfr(~ my_fun(.x, "H3"))
#>        X       Y Var1 Var2 df1_data$City
#> 1 001001  221001   H5   H3        Mexico
#> 2 001002  221044   H5   H3        London
#> 3 001003  221044   H5   H3            NY
#> 4 001004  221004   H4   H3         Otawa
#> 5 001005  001005   H3   H3     Liverpool

^{由reprex package (v2.0.1) 創建於 2022-04-14}