How can I define ALL my columns to be assigned as col_double()? The command below gives me a list of known columns to assigned. I just want ALL of them to be:
library(readr)
df<- read_delim("df.csv",
delim = ";", escape_double = FALSE,
col_types = cols(x1 = col_double()),
trim_ws = TRUE)
I have been trying to read CSV files with millions of rows and over 100 columns. I already managed to define all of them as numeric when I WRITE the CSV but, even so, when I try to read it back, it still gives logical columns.
Furthermore, I cannot rely on simply increasing the number of rows used to guess, and I would very much like to avoid converting columns after I read them.
My problem is probably much more basic, but I tried read_csv(), read_delim(), fread()... and I just don't know how to select all columns to be read as doubles.
You can set the default column type with the .default argument
library(tidyverse)
data = iris %>% select(-Species)
write_delim(data, "df.csv")
df <- read_delim("df.csv", col_types = cols(.default = col_double()))
sapply(df, class)