如何使用唯一 ID 使用循环代码构建子集

Question

I try to write a looping code with ID in a data.frame df . what I did right now is I build another list dm which contains the unique ID from df$ID :

dm<-df %>% select(ID) %>% unique()
for (i in 1:length( dm$ID)){
   df_new<-df %>% filter(ID %in% dm$ID[i]) 
...

Current codes can do what I need. But I wonder whether there is another way to do it without building dm? I want to build subset by each ID in df. Any suggestion?

Answer 1

Instead of looping over the unique 'ID' and subset ing, a faster option is split which will split the data.frame into list of data.frame based on the unique values of 'ID'

df_list <- split(df, df$ID)     

From here, we can either use lapply or a for loop

pdf(paste0(out_dir, output_date,'.pdf')) 
for(i in seq_along(df_list)) {
     ggplot(data = df_list[[i]]) + 
           ...
    } 
dev.off()

Or with lapply

pdf(paste0(out_dir, output_date,'.pdf'))    
lapply(df_list, function(dat) 
        ggplot(data = dat) + 
           ...
    )
dev.off()        

---

Regarding the creation of an object of unique 'ID', a better option is

for(un in unique(df$ID)) {
       df_new <- df %>%
                    filter(ID == un)
       ggplot(df_new) + 
           ...
  }