Question

我的列表有很多行但只有一列; 107,114,142,143,146。我想创建另一个列，具体取决于它是否具有特定值，请说114。

查看另一篇文章，我尝试了以下内容;

e_list$completed_forms <- rep(0, nrow(event_list))
e_list[e_list$e_list == 114, ][, "completed_forms"] <- "1"

数据列表

structure(list(event_list = c("211,202,214,240,104,105,106,107,114,117,118,139,140,141,142,143,146", 
"211,202,214,240,104,105,106,107,114,117,118,121,139,140,141,142,143,146", 
"211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146", 
"211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146", 
"211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146", 
"211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146", 
"211,202,214,240,215,216,104,105,106,107,114,117,118,120,121,139,140,141,142,143,146", 
"211,202,214,240,215,216,104,105,106,107,114,117,118,120,121,139,140,141,142,143,146", 
"211,114,117,118,146", "211,104,114,117,118,121,146", "211,202,104,114,117,118,121,139,141,142,143,146", 
"211,202,214,104,105,106,107,114,117,118,121,139,141,142,143,146", 
"211,202,214,215,104,105,106,107,114,117,118,121,139,141,142,143,146", 
"211,202,214,215,216,104,105,106,107,114,117,118,120,121,139,141,142,143,146", 
"211,202,214,215,216,203,240,104,105,106,107,114,117,118,120,121,139,140,141,142,143,146", 
"", "211,114,117,118,146", "211,114,117,118,146", "211,104,114,117,118,121,146", 
"211,202,104,114,117,118,121,139,141,142,143,146", "211,202,214,104,105,106,107,114,117,118,121,139,141,142,143,146", 
"211,202,214,215,104,105,106,107,114,117,118,121,139,141,142,143,146", 
"211,202,214,215,216,104,105,106,107,114,117,118,120,121,139,141,142,143,146", 
"211,202,214,215,216,217,240,104,105,106,107,114,117,118,120,121,139,140,141,142,143,146", 
"211,202,214,215,216,217,240,203,104,105,106,107,114,117,118,120,121,122,139,140,141,142,143,146", 
"211,202,214,215,216,217,240,203,104,105,106,107,114,117,118,120,121,122,139,140,141,142,143,146", 
"211,202,214,215,216,217,240,203,104,105,106,107,114,117,118,120,121,122,139,140,141,142,143,146", 
"", "211,114,117,118,146", "211,114,117,118,146"), lead_completed_new_forms = c(0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0)), .Names = c("event_list", "completed_forms"
), row.names = c(NA, 30L), class = "data.frame")

但我收到此错误并且我不确定原因，因为我已经在第一行的每一列添加了0。

[。data.frame（ tmp ，＆＃34; completed_forms＆＃34;，值）出错 =＆＃34; 1＆＃34;）：替换有1行，数据有0

Answer 1

整件事

library(stringr)
library(tidyverse)

df2 <- as.data.frame(str_split_fixed(df$event_list, ",", max(length(strsplit(df$event_list, ",")))))

names(df2) <- paste0("event_", 1:ncol(df2))

df2$sequence <- paste0("seq_", 1:nrow(df2))

df3 <- df2 %>% 
      gather(event, event_num, -sequence) %>%
      replace(. == "", NA) %>%
      filter(!is.na(event_num)) %>%
      select(-event) %>%
      mutate(Count = 1) %>%
      spread(event_num, Count) %>%
      replace(is.na(.), 0)

细分：

在标记这些数据集之前，我会做一些事情来清理这个数据集。这将使操作更容易。

head(df)
                                                                   event_list completed_forms
1         211,202,214,240,104,105,106,107,114,117,118,139,140,141,142,143,146               0
2     211,202,214,240,104,105,106,107,114,117,118,121,139,140,141,142,143,146               0
3 211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146               0
4 211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146               0
5 211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146               0
6 211,202,214,240,215,104,105,106,107,114,117,118,121,139,140,141,142,143,146               0

我将从使用str_split_fixed包中的stringr开始。如果您没有，install.packages(stringr)。这会将我们的dataframe转换为matrix，因此我们希望将其全部包含在as.data.frame()中。

library(stringr)

df2 <- as.data.frame(str_split_fixed(df$event_list, ",", max(length(strsplit(df$event_list, ",")))))

这样做需要一个像"134, 444, 105, 106"这样的字符串，并将它从一列拆分为4.我们将它拆分成的列数将是max(length())。这意味着如果我们有一个包含4个事件的列，一个包含14个事件，另一个包含44个事件，则它会将每列拆分为44列以容纳最大的列。

<强>检查：

> head(df2)
   V1  V2  V3  V4  V5  V6  V7  V8  V9 V10 V11 V12 V13 V14 V15 V16 V17 V18 V19 V20 V21 V22 V23 V24 V25 V26 V27 V28 V29 V30
1 211 202 214 240 104 105 106 107 114 117 118 139 140 141 142 143 146                                                    
2 211 202 214 240 104 105 106 107 114 117 118 121 139 140 141 142 143 146                                                
3 211 202 214 240 215 104 105 106 107 114 117 118 121 139 140 141 142 143 146                                            
4 211 202 214 240 215 104 105 106 107 114 117 118 121 139 140 141 142 143 146                                            
5 211 202 214 240 215 104 105 106 107 114 117 118 121 139 140 141 142 143 146                                            
6 211 202 214 240 215 104 105 106 107 114 117 118 121 139 140 141 142 143 146

让我们将标题变成事件编号来清理标题

names(df2) <- paste0("event_", 1:ncol(df2))

现在让我们创建一个列来标记所有这些我将称之为“序列”的事件序列。

df2$sequence <- paste0("seq_", 1:nrow(df2))

现在我们想将它从宽格式转移到长格式以便更容易消化。我们可以使用gather中的tidyverse函数执行此操作。我们还想将空格转换为NA值，然后将其从数据框中删除。 *此后我们可能也不需要event列。然后我会添加一些步骤将其重新转换为一个网格，其中列标题是事件编号，并且像您要求的那样具有1或0。

library(tidyverse)

df3 <- df2 %>% 
      gather(event, event_num, -sequence) %>%
      replace(. == "", NA) %>%
      filter(!is.na(event_num)) %>%
      select(-event) %>%
      mutate(Count = 1) %>%
      spread(event_num, Count) %>%
      replace(is.na(.), 0)

希望这能回答你的问题！

新列值以另一列为条件

1 个答案: