Why is as.Date slow on a character vector?

Question

I started using data.table package in R to boost performance of my code. I am using the following code:

sp500 <- read.csv(\'../rawdata/GMTSP.csv\')
days &         


        

Answer 1

This is an old question, but I think this tiny trick it might be useful. If you have multiple rows with the same date, you can do

data[, date := as.Date(date[1]), by = date]

It's much faster since it only processes each date once (in my dataset of 40 million rows it goes from 25 seconds to 0.5 seconds).