R语言中无法将原float类型变量转为double类型的求助

Hey there! Sounds like you’re stuck on converting those percentage columns to double type—let’s break this down and get it sorted.

The Root Issue #

The reason setting col_types = cols() to float or double isn’t working is almost certainly because your percentage columns have non-numeric characters (like the % symbol) in them. Readr can’t parse those into numeric types directly, which is why only character works.

Step-by-Step Solution #

Here’s how to fix this, using tidyverse tools since they’re great for data cleaning:

1. First, read the column as character (you already know this works, so stick with it for now):

   library(readr)
   library(stringr)
   
   df <- read_csv("your_dataset.csv",
                  col_types = cols(
                    your_percent_column = col_character()
                  ))
   

2. Clean the percentage symbol and convert to double
   You have two easy ways to do this:

   • Use str_remove() to strip the % then convert with as.numeric() (which gives you double-precision numbers in R):

     df_cleaned <- df %>%
       mutate(your_percent_column = str_remove(your_percent_column, "%") %>% as.numeric())
     

   • Or use parse_number() from readr—it automatically extracts numeric values from strings, ignoring non-numeric characters like %:

     df_cleaned <- df %>%
       mutate(your_percent_column = parse_number(your_percent_column))
     

3. Handle edge cases (if needed)
   If your data has other weird characters (like spaces, commas for thousands separators), expand the cleaning step:

   df_cleaned <- df %>%
     mutate(your_percent_column = str_remove_all(your_percent_column, "[%, ]") %>% as.numeric())
   

Bonus: Do It All in One Step #

You can even combine reading and cleaning into a single pipe to keep your code clean:

df_cleaned <- read_csv("your_dataset.csv",
                       col_types = cols(your_percent_column = col_character())) %>%
  mutate(your_percent_column = parse_number(your_percent_column))

Once you’ve done this, your column will be a double type, ready for any percentage calculations you need to run!

内容的提问来源于stack exchange，提问作者Reeza